Physics and Chemistry of Biological systems

SBP Group Seminar History

Date	Room	Speaker	Title
Jul 12 2024 11:30	Room 132	prof. Sebastian Springer (Constructor University, Bremen)	Redesigning MHC class I proteins for tumor immunotherapy
ABSTRACT: Killer T cells of the immune system can destroy virus-infected and tumor cells. These aberrant cells can be recognized because at their surface, they have MHC (major histocompatibility complex) class I proteins that display peptide fragments from the inside of the cell. My group is interested in all aspects of MHC class I proteins, especially their biochemistry, i.e., folding and ligand binding. Recombinant MHC class I proteins have a biotechnological role: they bind to tumor- or virus-specific T cells and can be used to detect, isolate, and activate them. We have used molecular dynamics simulations (in collaboration with Martin Zacharias in Munich) to introduce mutations in MHC class I proteins that make them more suitable for T cell detection: especially, we have generated peptide-empty forms that can shorten the production time from weeks to seconds. We are now looking to introduce further modifications, expand the approach to other proteins, and also screen for small molecules that support the stable conformation of the proteins. For this, we are open for collaboration.
Mar 23 2020 14:00	Webinar	Pierre Ronceray (Princeton, USA)	Active forces and stresses in living matter
ABSTRACT: A key feature of living systems is their ability to consume chemical energy to actively generate the forces they use to move and change shape. These forces are typically generated at the nanometer scale by motor proteins, and transmitted to larger scales by networks of fibers. I will first discuss the transmission of these active forces through the cell cytoskeleton and the extracellular matrix. I will show how the nonlinear mechanical properties of these biopolymers crucially affect force transmission by selecting and amplifying contractile stresses. We experimentally confirm these results using a novel stress measurement technique, Nonlinear Stress Inference Microscopy. In a second part, I will discuss how active forces emerge from Brownian noise at the sub-micron scale. From an observer's point of view, there is a fundamental bound to the amount of information that can be recovered by monitoring the dynamics of such systems. I will propose a practical method, Stochastic Force Inference, that efficiently utilizes this limited information to reconstruct force fields and infer dissipative currents in Brownian systems.
Mar 06 2020 14:00	Webinar	Rhoda Hawkins (Sheffield, UK)	Modelling the cell cytoskeleton
ABSTRACT: The mechanics of biological cells is dominated by the polymer protein network known as the cytoskeleton. In this talk I will outline ways in which we model the cytoskeleton using both analytical theory and simulations in combination with experimental data. My goal is to obtain a more profound understanding of the physics of this fascinating, complex, active matter. Furthermore we seek to comprehend how this out of equilibrium material not only affects cell mechanics but also drives cell shape change and movement. I will focus on two examples, both involving the cytoskeletal protein actin. The first involves stochastic simulations of force generation by actin polymerisation in phagocytosis, which is a biological process important in the immune system. The second involves continuum hydrodynamic models of contractile actomyosin and its role in cell migration in complex confining environments.
Mar 04 2020 14:00	Webinar	Matteo Paoluzzi (CNR, Rome)	Collective cell migration and glassy dynamics in simple models of biological tissue
ABSTRACT: Collective cell migration plays an important role in many biological processes ranging from embryogenesis to metastasis invasion. Some of these collective behaviors take place in situations where cells are at high packing fraction, as in the case of cell rearrangements in epithelial monolayers. Using the Voronoi tessellation of the cell center positions, Voronoi models describe a confluent tissue monolayer as a network of polygons. In self-propelled Voronoi models, a density-independent rigidity transition is driven by single-cell anisotropy and cell motility. We show that within Voronoi models collective cell migration is driven by minimal mechanical interactions at the level of a single cell. Moreover, Voronoi models show also heterogeneous rearrangements compatible with those observed in experiments and analogous with dynamical heterogeneities in colloidal glasses.
Mar 02 2020 14:00	Webinar	Abhinav Sharma (IPF, Dresden, Germany)	Lorentz forces induce inhomogeneity and flux in active systems
ABSTRACT: In contrast to externally driven systems, active matter has the hallmark of being driven out of equilibrium without breaking a spatial symmetry. Besides the application to biological systems, active matter serves as a paradigm to study the effect of broken time-reversal symmetry and nonequilibrium steady states in general. Much progress has been made in the understanding of the properties of active matter by using active Brownian particles (ABPs) as a model system. ABPs violate time-reversal symmetry by consuming fuel to generate motion, often referred to as self-propulsion. In this study we show that in the presence of a space- dependent magnetic field, a macroscopic flux emerges from a flux-free system of ABPs. This stands in marked contrast with similar phenomena in condensed matter such as the classical Hall effect, which requires an explicitly broken symmetry: a macroscopic velocity vector in addition to the symmetry breaking due to the magnetic field vector. We further demonstrate that passive tracer particles can be used to measure the essential effects caused by the Lorentz force on the active particle bath, and we discuss under which conditions this diffusive Hall-like effect might be observed experimentally.
Feb 14 2020 14:30	Room 005	Pablo Sartori (Rockfeller, USA)	Physical principles of protein complex assembly
ABSTRACT: In order to carry out cellular functions, proteins assemble into multi-component structures referred to as protein complexes. The properties of these complexes are fundamentally different from those of inert systems, which suggests the need for a new class of models. I will introduce a minimal model of protein mixtures and describe its four phases. Two phases are unlike those of conventional materials, one of which exhibits properties akin to cellular complexes. I will then discuss the need for a theoretical and experimental phenomenology of non-equilibrium assembly kinetics.
Feb 10 2020 14:00	Room 005	Farshid Jafarpour (UPenn, USA)	From Single Cell Phenomenology to Bacterial Population Dynamics
ABSTRACT: Genetically identical bacterial cells, even in identical environments, exhibit significant variability in their phenotypic behavior, such as their growth rates, division sizes, and generation times. With recent advances in single-cell technologies, we now can measure not only the distributions of these quantities but also the correlations between these variables both within and across generations. These statistical descriptions have paved the way for more accurate models of cellular growth and division. In this talk, I will discuss how the details of these new phenomenological models, such as the distributions of single-cell growth rates and the mechanism of cell-size control, affect various population-level quantities, in particular, the steady-state population growth rate, the rate of approach to the steady state, and the rate of genetic drift.
Nov 20 2019 14:00	Room 005	Yuanhua Huang ()	Bayesian methods for single-cell genomics: splicing and genetic variants
ABSTRACT: Single-cell RNA-seq (scRNA-seq) provides a comprehensive measurement of stochasticity in transcription, but the limitations of the technology have prevented its application to dissect variability in RNA processing events such as splicing. Here, I will present BRIE (Bayesian regression for isoform estimation) that resolves these problems by learning an informative prior distribution from sequence features, hence yields accurate estimates of exon inclusion ratios in single cells. I will also discuss ongoing extension of this model to incorporate cell-level features for enhancing splicing estimation and detecting associated splicing events. If time permits, I will briefly introduce our recent work on modelling of genetic variants in scRNA-seq data.
Nov 11 2019 16:00	Room 132	Korbinian Liebl (TUM Munich)	Simulation Studies and Force-field Development on DNA
ABSTRACT: DNAs sequence dependent structure and deformability is central to many biological functions, such as protein-binding, transcription and nucleosome packaging. Employing all-atom Molecular Dynamics simulations, we have investigated the coupling of DNAs helical parameters, the non-uniform elastic energy distribution along a DNA molecule under torsional stress and the stereo-chemical role of methyl-groups. While most simulations on regular double-stranded B-DNA structures are in good agreement with experimental parameters, we find that current force field descriptions over-stabilize dsDNA compared to single-stranded unstacked DNA. Thus, we have started building an entirely new DNA force field based on quantum-mechanical calculations. In this talk, I will present our approach to derive force-field parameters from ab initio calculations, show performance of our current force-field version and discuss further refinement possibilities.
Oct 10 2019 11:00	Room 005	Prof. G. Pavan (Politecnico di Torino, Department of Applied Science and Technology, Corso Duca degli Abruzzi 24, 10129 Torino, Italy)	Computational Approaches Toward Bioinspired Dynamic Materials
ABSTRACT: Many natural materials express fascinating dynamic properties and complex functionalitiesthat are impossible for common technological materials. These are supramolecular polymers (e.g., fibers, tubes, vesicles, etc.) built via the self-assembly of fundamental building blocks such as proteins, lipids, peptides, to name a few. dynamic, adaptive and stimuli-responsive propertiesby the experiments. Learning a priori how to design artificial materials with similar according to the similar self-assembly principles would be a breakthrough in many fields. However, thedesign rules to control such bioinspired dynamic systems are prohibitively difficult to catchWe combine multiscale molecular models (atomistic and coarse-grained ),advanced simulation approaches and machine learning to access the intrinsic dynamicsand the dynamic properties of supramolecular assemblies at a sub-molecular resolution. This permits us, for example, to study the molecular factors that control how much and how fast/slow an assembly responds to specific stimuli, and to investigate at unprecedentedresolution how complex self-assembled systems behave, or evolve, out-of-equilibrium. The scientific advance that can be obtained holds a great potential toward the rational designof next-generation dynamic materials for various technological applications.
Jul 30 2019 14:00	Room 132	Dr. Jacopo Grilli (Quantitative Life Sciences, ICTP, Trieste)	Laws of diversity and variation in microbial communities
ABSTRACT: How coexistence of many species is maintained is a fundamental and unanswered question in ecology. Coexistence is a puzzle because we lack a quantitative understanding of the variation in species presence and abundance. Whether variation in ecological communities is driven by deterministic or random processes is one of the most controversial issues in ecology. I will consider the variation of species presence and abundance in microbial communities from a macroecological standpoint. We identify three novel, fundamental, and universal macroecological laws that characterize the fluctuation of species abundance across communities and over time. These three laws in addition to predicting the presence and absence of species, diversity and other commonly studied macroecological patterns allow testing mechanistic models and general theories aiming at describing the fundamental processes shaping microbial community composition and dynamics. I will conclude by showing that a mathematical model based on environmental stochasticity quantitatively predicts the three macroecological laws, as well as non-stationary properties of community dynamics.
Jul 16 2019 12:00	Room 005	Prof. Daniel Remondini (Bologna University, Italy)	Extracting information from DNA sequences through stochastic modelling
ABSTRACT: DNA sequences convey information on protein codification, but recent experimental observations point at active roles also on spatial conformation (chromatin) through epi-genetic mechanisms. A key question is: which information about DNA regulation is embedded into DNA sequence? We show recent results based on our analyses, in which we consider a DNA sequence as the realization of a stochastic process combined with data analysis methods.
Jul 15 2019 14:30	Room 005	Prof. Nicola Neretti (Brown University, USA)	Genomic and Epigenomic Instability in Cellular Senescence and Aging
ABSTRACT: This talk will describe how epigenomic changes that occur in aging cells affect genomic stability and the transcription of genes and other genomic elements. Senescent cells, in particular, undergo dramatic alterations to their chromatin landscape that affect genome accessibility and their transcriptional program. These include the loss of DNA-nuclear lamina interactions, the distension of centromeres, and changes in chromatin composition that can lead to the activation of retrotransposons. I will show some recent results on the use of cell-free DNA to detect some of these changes in vivo through liquid biopsies. I will describe how senescent cells undergo a profound reconfiguration of the 3-dimensional structure and organization of chromosomes that is associated with the activation of the senescent transcriptional program. This led us to develop novel methods to infer the 3D structure of chromosomes from chromatin conformation capture (3C) data. Finally, I will show how we are using single cell transcriptomics to characterize heterogeneity in senescent cells, as well as mechanisms of secondary senescence and differential response to senolytics.
Jul 08 2019 14:00	Room 128	Prof. Sharon Ruthstein (Department of Chemistry, Faculty of Exact Sciences, Bar Ilan University)	The tale of the cellular copper cycle
ABSTRACT: Copper's ability to accept and donate single electrons makes it an ideal redox cofactor, and thus one of the most essential metal ions to the survival of the cell. However, copper ions are also involved in the Fenton reaction and hence capable of driving the generation of deleterious hydroxyl radicals, which are deleterious to the cell. Hence, both prokaryotic and eukaryotic systems have developed a considerable regulation mechanism to maintain negligible copper concentration, in the femtomolar concentration. Deciphering this regulation mechanism in eukaryotic and prokaryotic systems is tremendously important from several reasons: first, it will assist in developing new therapeutic agents that will control the in-cell copper concentration. Second, copper has been used throughout much of the human civilization as an antimicrobial agent, and understanding its cellular pathway can lead to development of new generation of antibiotics. In this talk, I will shed light on several important copper regulation systems in the human cell and in E.coli: the human Ctr1-Atox1-ATP7b pathway, the copper periplasmic efflux system, CusCFBA, and the Cu(I) metal sensor, gene expression regu-lation system, CueR. Using Electron Paramagnetic Resonance (EPR) spectroscopy, along with biochemical and computational work, I will present structural models for Ctr1-Atox1-ATP7B system, CusB and CueR in the apo and functional state. Then, based on the structural constraints and cell data I will explain their mechanism of action. Last, I will demonstrate how basic understanding of the function of these systems can assist us in designing new class of biomarkers and antibiotics.
Jun 05 2019 14:00	Room 132	Nikola Minovski (National Institute of Chemistry, Ljubljana)	QSAR & The Evolution of Dimensionality Simulating Induced-fit Mechanisms in Drug Discovery
ABSTRACT: One of the earliest, but still popular and useful computational strategies in the modern drug discovery is assuredly the methodology of establishing a quantitative relationship between the chemical structures for a given compounds class and their experimentally-determined biological activity values, i.e., an in silico modeling (statistical) technique commonly known as quantitative structure-activity relationship (QSAR). When QSAR as a paradigm was introduced for the first time in its entirety in the middle of the previous century, a plethora of QSAR approaches have been devised. Except the traditional topological 2D-QSAR methods (e.g., the popular Free-Wilson and Hansch-Fujita models), which are still very useful, a further breakthrough in the QSAR progress was the introduction of 3D-QSAR methods. Notwithstanding their relatively high success in the identification of promising hit compounds in general, a major drawback of these methodologies is the lack of dynamic nature of the systems describing the binding of a ligand to the biological target (induced-fit phenomena). This lecture will be a brief introduction to the evolution of dimensionality beyond the 3rd dimension in the QSAR modeling with a special emphasis on the simulation of ligand-receptor induced-fit phenomena and its impact on the accuracy of ligands binding affinity predictions.
May 30 2019 14:00	Room 132	Prof. Aniket Bhattacharya (University of Central Florida)	DNA folds inside a nanochannel: scaling and non-equilibrium dynamics
ABSTRACT: DNA is one of the most important biomolecules in living organism, forms a helix from two intertwined strands with complementary base pairs. Biological functions of a DNA depend on its mechanical properties, which in turn depend on its sequence specificity. Under physiological condition a double stranded DNA (dsDNA) is described as a semi-flexible biopolymer with persistence length of 50 nm, while a single-stranded DNA (ssDNA) is quite flexible. Recently straightening a dsDNA inside a nanochannel is being explored as an alternate method to determine DNA sequences at a single molecule level without replication. First, I will present coarse grained (CG) models for fast computations of DNA conformations and dynamics. I will use scaling arguments validated by Brownian dynamics (BD) simulation results performed on the CG models to demonstrate how the equilibrium DNA conformations change inside a nanochannel, as one varies the persistence length (stiffness) and the channel width [1, 2]. I will then show the transients and the steady states of an initially straightened DNA inside a nanochannel squeezed by a Nano-dozer assay. I will compare the time dependent density profiles from the BD simulation with those obtained from a Nonlinear Partial Differential Equation (NPDE) approach recently introduced by Khorshid et al., and demonstrate how this combined approach can be effectively used to study nonequilibrium dynamics of very long dsDNA segments inside a nanochannel [3]. For stiff chains in nanopores, we further show that chain compression proceeds through a unique folding kinetics driven by repeated double fold nucleation events and growth of nested folds. We show that the folding kinetics can be understood by coupling a theory for deterministic contour spooling across the folds with a dynamically varying energy landscape for fold nucleation. These findings are critical for understanding compression of nanochannel confined DNA in the sub-persistence length (Odijk) regime [4]. (*) Work done in collaboration with Kurt Binder, H.-P. Hsu, Aiqun Huang, and Walter Reisner [1] Aiqun Huang and Aniket Bhattacharya, DNA confined in a two-dimensional strip geometry, Euro Phys. Lett. 106, 18004 (2014). [2] Aiqun Huang, H.-P Hsu, Aniket Bhattacharya, and Kurt Binder, Semiflexible macromolecules in quasi-one-dimensional confinement: Discrete versus continuous bond angles, J. Chem. Phys. 143, 243102 (2015). [3] Aiqun Huan, Walter Reisner, and Aniket Bhattacharya, Dynamics of DNA Squeezed inside a Nanochannel via a Sliding Gasket, Polymers 2016, 8 (10), 352; [4] Simon Bernier, Aiqun Huan, Walter Reisner, and Aniket Bhattacharya, Evolution of Nested Folding States in Compression of a Strongly Confined Semiflexible Chain Macromolecules 2018, 51 (11), 40124022
Feb 28 2019 14:00	Room 135	Dott. Aldo Glielmo (Kings College London)	Gaussian process models for force fields and wave functions
ABSTRACT: In the last decade there has been a proliferation of successful applications of Machine Learning (ML) algorithms to complex problems in physics, a good example being the use of Gaussian Process (GP) regression to approximate interatomic force fields at the DFT level of accuracy [1]. The quality of a GP regression is entirely dictated by the quality of its kernel function, which encapsulates the prior knowledge on the system studied. I will present a general scheme to build GP kernels that fully encode physical prior information while maintaining a controllable level of complexity [2]. The procedure extends and clarifies well known previous approaches (e.g. [3] and [4]) and lies in a systematic expansion of the approximating force field into n-body interactions via the design of n-body kernels. The problem of choosing the order n, best suited to describe a given system can be then approached in a principled way through Bayesian model selection. GP predictions coming form the chosen n-body kernels can further be mapped exactly onto explicit basis functions giving rise to non-parametric classical potentials. These are as fast as standard classical potentials [5] (and thus crucially orders of magnitude faster than typical ML force fields), but provide substantially lower errors with respect to DFT forces. Furthermore, they can be generated automatically with no need of complex non-linear parametrisation and optimisation. [1] Bartók et al. (2010). Physical Review Letters, 104(13), https://doi.org/10.1103/PhysRevLett.104.136403 [2] Glielmo et al. (2018). Physical Review B, 97(18), http://doi.org/10.1103/physrevb.97.184307 [3] Bartók et al. (2013). Physical Review B, 87(18), http://doi.org/10.1103/PhysRevLett.104.136403 [4] Glielmo et al. (2017). Physical Review B, 95(21), http://doi.org/10.1103/PhysRevB.95.214302 [5] Zeni et el. (2018). The Journal of Chemical Physics, 148(24), http://doi.org/10.1063/1.5024558
Feb 27 2019 10:00	Room 132	Dr. Vasyl Mykuliak (Tampere University, Finland)	Stable 3-helix intermediates revealed in talin and α-catenin during mechanical unfolding
ABSTRACT: In addition to chemical stimuli, protein functions can be modulated by mechanical signals. Mechanical stability is a key feature for the regulation of functions of structural scaffolding proteins, such as talin and α-catenin. Talin physically connects intracellular actin filaments to extracellular matrix via transmembrane integrin. Similarly, α-catenin provides connection between cell-cell adhesions and the cytoskeleton. Talin and α-catenin undergo unfolding under mechanical load, what opens buried binding sites for protein partners. We used a combination of atomistic steered molecular dynamics (SMD) simulations and single-molecule atomic force microscopy (smAFM) to study how talin rod subdomains and α-catenin domains unfold. SMD and smAFM reveal that all studied proteins unfold through stable 3-helix intermediates. The 5-helix bundles were most stable in our experiments, while the 4-helix bundles were easily unfolded to 3-helix intermediates. We hypothesize that the 3-helix intermediates are distributed within talin rod during unfolding and may serve as binding epitopes for protein partners.
Feb 18 2019 16:00	Room 132	Dr. Michele Allegra (Institut de Neurosciences de la Timone, Marseille)	Clustering by the local intrinsic dimension: the hidden structure of real-world data
ABSTRACT: It is well known that a small number of variables is often sufficient to effectively describe high-dimensional data. This number is called the intrinsic dimension (ID) of the data. What is not so commonly known is that the ID can vary within the same dataset. This fact has been highlighted in technical discussions, but seldom exploited to gain practical insight in the data structure. We developed an approach to cluster regions with the same local ID in a given data landscape. Surprisingly, we find that many real-world data sets contain regions with widely heterogeneous dimensions. These regions host points differing in core properties: folded vs unfolded configurations in a protein molecular dynamics trajectory, active vs non-active regions in brain imaging data, and firms with different financial risk in company balance sheets. Our results show that a simple topological feature, the local ID, is sufficient to uncover a rich structure in high-dimensional data landscapes.
Feb 11 2019 16:00	Room 132	Dott.ssa Barbensi (Oxford University)	Grid diagrams as tools to investigate knots space and the unknotting function of type II topoisomerases
ABSTRACT: The action of type II topoisomerases on covalently closed DNA molecules can change their topology, resulting in a range of different knot types. Grid diagrams are a computationally convenient way to represent knot types and local deformations between them. We use a grid diagrams based model to investigate topological consequences of intersegmental passages occurring in circular DNA molecules. We extend and synthesise earlier investigations of DNA topoisomerase selection of sites of action and of knot adjacency in the knot space by looking at neighbouring subspaces in the graph of the configurations, modeled as a network of grid diagrams with increasing complexity. We suggest a grid-based calculation as a new and computationally convenient framework for investigating unbiased knotting probability biopolymers and to study the role of local juxtaposition geometry for Topoisomerases action. Joint work with Dorothy Buck, Heather A. Harrington and Daniele Celoria and Andrzej Stasiak
Jan 30 2019 14:00	Room 128-129	Dr. Luca Ghiringhelli (Fritz-Haber-Institut, Max-Planck Berlin)	Data-Driven Materials Science: the Critical Role of the Descriptor
ABSTRACT: The number of possible materials is practically infinite, while only few hundred thousands of (inorganic) materials are known to exist and for few of them even basic properties are systematically known. In order to speed up the identification and design of improved, new, and even novel optimal (functional) materials for a desired property or process, strategies for quick and well-guided exploration of the materials space are highly needed. A desirable strategy would be to start from a large body of experimental or theoretical data, and by means of (big-)data-analytics methods, to identify yet unseen patterns or structures in the data. This leads to the identification of maps (or charts) of materials where different regions correspond to materials with different properties. The main challenge on building such maps is to find the appropriate descriptive parameters (called descriptors) that define these regions of interest. Here, I will present a suite of artificial-intelligence methods, recently developed by us for the machine-aided identification of descriptors and materials maps. These methods are applied to, e.g., crystal-structure prediction, the metal/nonmetal classification, the prediction of novel 2D and 3D topological insulators, and the construction of a tolerance factor for the stability of perovskites, enabling the costless prediction of thousands of new candidate perovskites. I will also describe the infrastructure to perform such analyses online, via the "Big-data-analytics toolkit" within the framework of the Novel-Materials-Discovery (NOMAD) Laboratory.
Jan 28 2019 12:00	Room 128-129	Prof. M. Caselle (University of Torino)	Simple regulatory strategies to control stochastic fluctuations in gene expression
ABSTRACT: Fine tuning of gene expression is essential for many cellular processes, from cellular responses to external stimuli to the cell cycle and circadian clocks. However, gene expression is subject to stochastic fluctuations, naturally inducing an uncertainty in the amount of gene products, with potential consequences on biological functions and phenotypes. In this talk, after a general introduction to gene expression and its regulation, we discuss a few regulatory strategies which guarantee optimal control of these stochastic fluctuations.
Jan 18 2019 10:00	Room 005	Dr. Daniele Granata (University of Copenhagen)	Exploring the impact of sequence on protein structure, dynamics and function
ABSTRACT: The extraordinary successes achieved in the last years by bioinformatics, structural biology, and functional studies of proteins have produced an unprecedented amount of data that holds promise to shed light on the workings of these fundamental cells molecular machines. In particular theoretical and computational approaches produced powerful methods to explore, piecewise, the renowned paradigm sequence-structure-function, rationalizing the microscopic determinants of protein folding and providing interpretations and testable hypothesis for experiments concerning protein dynamics and function. But is it possible to recapitulate the entire flow of information contained in the paradigm relying only on homologous sequences of the protein of interest? Here I will present the results from my previous investigations of the above paradigm, together with my future research strategies to understand how the analysis of sequence variation in a protein family can be used directly for the interpretation of the phenotypic effect of genome variability, i.e. how mutations in the protein sequence can impact its stability and function. Indeed based on statistical inference, data-mining and molecular modelling approaches, I plan to devise new bioinformatics tools for high-throughput screening and prediction of the mutational (or fitness) landscape of proteins in application to the diagnosis of genetic diseases, as well as in protein engineering and design.
Jan 11 2019 10:00	Room 128	Dr. Marcin J. Skwark (University of Cambridge)	Solving protein folding problem and bacterial evolution with statistical physics (and some machine learning)
ABSTRACT: Life and its operation is defined by interactions between partners. Protein structures are formed and stabilized by contacts between amino acids. The same, weak interactions make for protein complexes and determine specificity of binding in biological contexts. Observing these phenomena through direct experiments is often difficult and costly. Cheap, indirect observations (e.g. co-expression patterns, correlated mutations...), on the other hand, frequently result in misguided, low-precision results, due to confounding effects of transitive correlations. This talk discusses a recent revolution in computational biology, which by use of exponential family models of statistical physics (inverse Ising and Potts models; c.f. Lapedes, 1999) made high accuracy protein contact prediction and inference of other causal relationships possible. I will talk about some of my contributions to the field, both in terms of interplay between model, data, and inference method (c.f. Feinauer, Skwark et al., 2014; Skwark, Michel et al, 2014), as well as the search for optimal use of these methods in structural bioinformatics (c.f. Golkov, Skwark et al., 2016). Analogous models can be applied to infer epistatic interactions from multiple sequence alignments of bacterial genomes. Despite the problem being massively over-parametrized, we have shown (Skwark et al, 2017), that an ensemble of Potts models accurately captures evolutionary pressures in Streptococcus pneumoniae. These results closely recapitulate these of prior GWAS experiment (Chewapreecha et al, 2014), but in an entirely unsupervised manner. Nowadays, these methods have been widely embraced by the academia and industry. While there has been substantial progress in terms of inference efficacy, applications to other problems have yet to be fully successful. I will present therefore several open possibilities for the field to develop.
Jan 09 2019 10:00	Room 005	Dr. Toma Tebaldi (University of Trento)	Mapping post-transcriptional regulation by positional analysis of RNA sequencing data
ABSTRACT: RNA impacts nearly every aspect of gene expression. Thus, it comes as no surprise that several diseases are caused by alterations in RNA metabolism, affecting post-transcriptional processes such as RNA splicing and polyadenylation, RNA modification, RNA localization and translation. Coding and non-coding RNA molecules are dynamically coated by RNA-binding proteins to form ribonucleoprotein complexes. This intricate network of interactions regulates the fate of RNAs within cells. Physiological RNA-protein interactions are frequently altered in neurodegenerative diseases and in tumors, ultimately perturbing the protein levels of specific genes through mechanisms that are not entirely understood. To overcome this knowledge gap, the last few years have witnessed a rapid adoption of deep sequencing approaches, capable of studying in-vivo post-transcriptional regulation with unprecedented resolution. Among these, UV crosslinking and immunoprecipitation followed by deep sequencing (CLIP-Seq) identifies protein-RNA interactions with nucleotide resolution across the transcriptome. Ribosome profiling is instead a powerful technique to study translation at the transcriptome-wide level, generating unique information concerning ribosome positions along RNAs. I will present some case studies to show how developing positional analysis of RNA sequencing data with nucleotide resolution can be effectively used to unveil the role of RNA-protein complexes, non-coding RNAs and ribosomes, and their involvement in multiple and diverse pathologies such as neuronal degeneration and tumorigenesis.
Jan 08 2019 14:00	Room 005	Dr. Fabio Anselmi (IIT Genova & MIT Boston)	Towards a more interpretable deep learning and its relevance for physical science.
ABSTRACT: Major advances in artificial intelligence through deep learning methods are revolutionizing the way data are treated in many fields outside computer science like physics chemistry and astronomy. However these algorithms lack interpretability of the results, a key feature to develop new and understandable models of nature.I argue that this gap is caused by the fact that current architectures are severely under-constrained, lacking key model biases found in the statistics of the data. I show how learning such biases is fundamental to develop physically-informed and interpretable deep learning models of experimental data. I end giving potential applications of machine learning methods in the realm of physical science. Examples include data driven learning of: dynamical models of a physical system, symmetries in high energy physics, new molecules with specific chemical properties, automatic classification of galaxies shapes.
Dec 18 2018 16:00	Room 132	Dr. Barbara Capone (University of Roma III)	Coarse graining polymer solution: multiscale, scaling theories and applications
ABSTRACT:
Oct 01 2018 14:00	Room 132	Prof. Pietro Faccioli (University of Trento)	Coupling structural and electronic dynamics of biomolecules in non-equilibrium conditions
ABSTRACT: Time-resolved spectroscopy provides a very sensitive technique to probe the dynamics of nuclear and electronic degrees of freedom in biomolecules. Unfortunately, performing microscopic calculations of these spectrA is very hard, and this limits the amount of information which can be deconvolved from the experimental data. In particular, the structural dynamics is often dominated by rare transitions, which are notoriously hard to simulate even at the classical level. In addition, computing spectroscopic observables unavoidably requires to account for the quantum dynamics of the electronic excitations, in their open environment. In this seminar, I will review our recent progress toward developing a unified and viable approach to the non-equilibrium quantum and classical dynamics biomolecules, which is based on combining the Feynman-Vernon path integral formalism, variational schemes and second quantization techniques [1]. This way, it is possibile to develop approximations which make it computationally feasible to simulate transitions as complex as protein folding at the atomic level of resolution [2,3] and to make theoretical predictions for spectroscopic observables with can be directly compared with experimental data. In particular, I will focus on a recent application of this method to extract high resolution information about the protein folding mechanism from time-resolved near UV CD spectra [4]. Selected References: [1] E. Schneider, S. a Beccara, F. Mascherpa and P. Faccioli, Phys. Rev. B 94, 014306 (2016). [2] S. Orioli, S. a Beccara and P. Faccioli, J. Chem. Phys. 147, 064108 (2017) [3] S. a Beccara, L. Fant & P. Faccioli, Phys. Rev. Lett. 114, 098103 (2015) [4] A. Ianeselli et al., J.A.C.S 140 3674 (2018).
Sep 28 2018 15:30	Room 132	Dr. Luca Barberi (Univ. Paris-Sud, France)	Inferring the helix-helix electrostatic interaction strength from the structure of dense DNA toroids
ABSTRACT:
Jul 23 2018 15:00	Room 132	Dr. Mattia Bernetti (Università di Bologna, Dipartimento di Farmacia e Biotecnologie)	Accessing infrequent biomolecular processes through biased molecular dynamics simulations
ABSTRACT: Traditionally, the computational contribution to the drug discovery pipeline relies mostly on the prediction of ligand-target binding poses, mainly through molecular docking procedures. In this sense, essential ingredients for successful campaigns are a reliable representation and characterization of the biomolecules that have to be drugged.[1] On the one side, a crucial challenge is posed by emerging tough targets, as intrinsically disordered proteins (IDPs) are. Indeed, achieving structural description of these species is still a major issue. Additionally, well-recognized limitations associated with rigid treatment of the molecular targets represent a general problem for most docking protocols. In this context, ensuring full flexibility becomes a pivotal objective to asses binding features of promising ligands. From a computational standpoint, addressing such requirements calls for a dynamical representation, which is provided by molecular dynamics (MD) simulations.[2] Nevertheless, notwithstanding the remarkable advances that have been achieved in computer hardware over the last decades, conventional MD cannot yet easily access the involved timescales. Herein, we approached both these key points taking advantage of biased MD simulations. Specifically, as for the first aspect, we characterized equilibrium properties of a test-case IDP accounting on metadynamics coupled with currently available instruments, namely a force field non-specifically devised to handle disordered states.[3] Concerning the second point, we conceived a fully-flexible docking procedure based on scaled MD and validated it over two pharmaceutically relevant targets.[4] By reproducing full binding events starting from protein-ligand dissociated states, the methodology allowed to successfully reproduce experimentally determined crystal poses for all of the considered compounds. Despite some limitations, the results were encouraging and suggested promising potential to provide future support in a pharmaceutical perspective. REFERENCES Gioia, D.; Bertazzo, M.; Recanatini, M.; Cavalli, A. Dynamic Docking : A Paradigm Shift in Computational Drug Discovery. Molecules 2017, 22 (11), 121. De Vivo, M.; Masetti, M.; Bottegoni, G.; Cavalli, A. Role of Molecular Dynamics and Related Methods in Drug Discovery. J. Med. Chem. 2016, 59 (9), 40354061. Bernetti, M.; Masetti, M.; Pietrucci, F.; Blackledge, M.; Jensen, M. R.; Recanatini, M.; Mollica, L.; Cavalli, A. Structural and Kinetic Characterization of the Intrinsically Disordered Protein SeV NTAIL through Enhanced Sampling Simulations. J. Phys. Chem. B 2017, 121 (41), 95729582. Bertazzo, M.; Bernetti, M.; Recanatini, M.; Masetti, M.; Cavalli, A. Fully Flexible Docking via Reaction- Coordinate-Independent Molecular Dynamics Simulations. J. Chem. Inf. Model. 2018, 58 (2), 490500.
Jul 18 2018 14:30	Room 132	Dr. Francesco Colizzi (IRB (Barcelona))	Predicting the limit of intramolecular H-Bonding
ABSTRACT: The energetics of intramolecular recognition processes are governed by the balance of pre-organization and flexibility that is often difficult to measure and hard to predict. Here, by using state-of-the-art molecular dynamics simulations, we predict and quantify the effective strength of intramolecular interactions between H-bond donor and acceptor sites separated by a variable alkyl linkerin chloroform, water, and in crowded solutions. We observe a fine balance of entropic and enthalpic contributions that posits a solvent-dependent limit to the occurrence of intramolecular H-bonding. H-bond free energies are systematically shifted by ~13 kJ/mol between water and chloroform with a ~6 kJ/mol solvent-independent entropic penalty for the addition of a rotor. Molecular crowding shows little effects on thermodynamic equilibrium but it induces pronounced variations on H-bond kinetics by reducing water availability. The results showcase a general strategy to interrogate molecular interactions in different environments and extend the limits of current experiments towards the prospective prediction of H-bond interactions in pharmaceutical, agrochemical, and technological contexts.
Jul 03 2018 14:30	Room 132	Rosa Di Felice (Department of Physics and Astronomy, University of Southern California, Los Angeles and CNR-NANO-S3 Modena, Italy)	Elucidating binding mechanisms of class 2 CRISPR/CAs9 complexes by molecular dynamics simulations
ABSTRACT: The RNA-guided endonuclease Cas9 from microbial immune systems is a powerful tool for genome editing in eukaryotic cells. The nuclease activity of Cas9 can be triggered when there is imperfect complementarity between the RNA guide and target DNA. This is a problem in the context of genetic therapies, where one would like precise specificity. Crystal structures of bound and unbound Cas9 have revealed substantial information on CRISPR/Cas9 systems. However, resolved crystal structures have two major limitations: (i) the nuclease active site is mostly far from the target DNA (DNAt) strand cleavage site and (ii) there is lack of an intact non-target DNA (DNAnt) strand. Here I focus on the latter limitation. Using a working hypothesis formulated by Slaymaker and coworkers [1], based on the existence of a positively charged patch in Cas9 that ideally accommodates the negatively charged backbone of the unwound DNAnt strand, we modeled a longer DNAnt strand in the crystal structure of the ternary complex from PDB ID 4UN3. The molecular dynamics (MD) of this complex on the scale of 1.5 ?s, complemented by electron paramagnetic resonance (EPR) measurements of distances between spin labels attached to the backbone of DNAt and DNAnt, offers consolidation to this working hypothesis[2]. I present an overview of CRISPR/Cas9 operation and the results of the joint MD-EPR work[2] on DNAt-DNAnt distances, with implications on detecting the fate of the unwound DNAnt strand after binding of the RNA and endonuclease. Furthermore, I outline current ongoing efforts, preliminary results and future plans to unravel possible structural ways to improve the binding specificity of CRISPR/Cas9 complexes through mutations in the above mentioned positively charged patch [1]. [1] Slaymaker, I. M.; Gao, L.; Zetsche, B.; Scott, D. A.; Yan, W. X.; Zhang, F. Rationally engineered Cas9 nucleases with improved specificity. Science 2015, 351, 8488. [2] Tangprasertchai, N.; Zhang, X.; Di Felice, R.; Slaymaker, I. M.; Vazquez Reyes, C.; Jiang, W.; Rohs, R.; Qin, P. CRISPR-Cas9 mediated DNA unwinding detected using site-directed spin labeling. ACS Chem. Biol. 2017, 12, 1489-1493.
Mar 28 2018 14:00	Room 132	Prof. Xevi Biarnés (Laboratory of Biochemistry, IQS School of Engineering, Universitat Ramon Llull.)	Modelling and engineering carbohydrate active enzymes
ABSTRACT: Carbohydrate Active Enzymes (CAZYmes) catalyze the biosynthesis, breakdown and modification of glycan structures. This is a widespread family of proteins present in all kingdoms of life with diverse functions in cell structure, energy storage and cellular signaling. CAZYmes are used as biocatalysts to produce functionalized oligo- and polysaccharides, which give access to novel biomaterials for biomedical and biotechnological applications. In the perspective of using these and other enzymes as biocatalysts to produce non-natural compounds, the starting substrates are usually non-natural as well. Since the enzyme- substrate interactions are not naturally-optimized in such cases, it is expected to be much room for the improvement of the catalytic efficiency by controlling the substrate specificity. I will present here two examples of the application of structural bioinformatics and molecular modelling tools to assist in the design of protein engineering experiments of CAZYmes. The first example is based on a wide-genome analysis of chitin deacetylases (CDAs) sequences and structures combined with new crystallographic and experimental mutational data [1]. This analysis lead to the proposal of the "subsite capping model" to rationalize the diversity in product profiles of different CDAs [1,2]. Secondly, the engineering of a glycosyl hydrolase (GH) to revert its natural degrading activity into a synthetic enzyme will be presented [3]. A new computational algorithm (BindScan) was used as part of the protein engineering protocol. This algorithm exhaustively casts all the positions on a given protein sequence by individually mutating each position and measuring the effect on the binding affinity to a given compound. This information can then be used to design single point mutations or to guide directed evolution experiments for the improvement of substrate specificity in the working enzyme. References [1] Andrés E, Albesa-Jové D, Biarnés X, Moerschbacher BM, Guerin ME, Planas A (2014). Angewandte Chemie-International Edition, 53:6882-6887. [2] Aragunde H, Biarnés X, Planas A (2018). International Journal of Molecular Sciences, 19:412. [3] Bissaro B, Durand J, Biarnés X, Planas A, Monsan P, Fauré R, O'Donohue MJ (2015). ACS Catalysis, 5:4598-4611.
Mar 11 2018 14:30	Room 128-129	prof. Dante Chialvo (Center for Complex Systems & Brain Sciences - CEMSC3 - UNSAM - Buenos Aires, Argentina)	Ubiquity of criticality in neural function across scales: recent empirical findings, speculations an caveats
ABSTRACT: The organisation of biological form and function is a classic problem, cut-crossing disciplines, which include a variety of complex spatiotemporal patterns. Historically, work first focussed into understanding oscillations and, more recently, attention included scale-free collective fluctuations, some of them shown to correspond to critical phenomena. In this lecture we will review our work across several scales characterising such phenomena in brain function, proteins and mitochondria networks.
Mar 02 2018 14:00	Room 005	Dr. Andrea De Martino (CNR)	Inverse-modeling bacterial growth: how do growth and gene expression lead to fitness?
ABSTRACT: Over the past decade, novel experimental techniques have probed proliferating bacteria at unprecedented detail, revealing how growth physiology is shaped by an extended crosstalk with gene expression. In a broad perspective, such results challenge the traditional idea of optimality associated to exponentially growing cells. To forego this assumption, we use a data-driven approach to infer the distribution of metabolic phenotypes of exponentially growing E. coli cells from different empirical inputs, including bulk and single-cell measurements. A quantitative "rate-distortion curve" is generically found to link the growth rate to the (computable) strength of metabolic regulation required to achieve it. In particular, inferred phenotypes can be analyzed relative to this bound, providing insight into what, if anything, is being (approximately) optimized by experimental populations. A deeper look into the physical meaning of the "inverse temperature" that controls inferred distributions shows surprising connections to dynamics. Specifically, the initial size and previous history of a colony may be crucial proxies to determine how an exponentially growing population organizes in the phenotypic space. New experiments on cancer cells strongly support this scenario.
Jan 19 2018 14:00	Room 004	Dr. Michele Allegra ()	Synthetic charts for navigating high-dimensional data spaces
ABSTRACT: Maps are ways of representing the salient features of a manifold, typically a two-dimensional surface. Unlike modern maps, which aim at faithfully representinging a surface, stick charts used by ancient Marshall island peoples achieved a synthetic, compact representation of an ocean area with essential information about the main islands and sea swells. Similarly, data analysis in high-dimensional spaces must aim at obtaining a synthetic description of a data set revealing its main structures, without attempting to provide a complete description. In this lecture I will review some recent results on the problem of charting high-dimensional data spaces, focusing on a method for finding manifolds of different intrinsic dimension within a dataset.
Jan 15 2018 14:00	Room 004	Dr. Michele Allegra ()	Wine out of turnips: finding changes in coherent brain activity patterns with coherence density peak clustering
ABSTRACT: In principle, functional magnetic resonance imaging (fMRI) data allow observing large-scale cortical activity patterns associated to a given cognitive task. In practice, retrieving these patterns from the data in a reliable way is a highy non-trivial endeavour. We approached this problem though a clustering method developed in house (coherence Density Peak clustering). In particular, we applied the method to data from a task with alternative strategies, where we could reveal pattern changes associated to learning and strategy change. In this lecture, I will look back at the problems and difficulties encountered in this project, and the mistakes we made, which taught us some general lessons in data analysis.
Nov 27 2017 16:00	Room 128	Giovanni Petri (ISI Foundation, Torino)	Structure and evolution of topological brain scaffolds
ABSTRACT: Topology is one of the oldest and more relevant branches of mathematics, and it has provided an expressive and affordable language which is progressively pervading many areas of biology, computer science and physics. I will illustrate the type of novel insights that algebraic topological tools are providing for the study of the brain at the genetic, structural and functional levels. Using brain gene expression data, I first will construct a topological genetic skeleton, together with an appropriate simplicial configuration model, pointing to the differences in structure and function of different genetic pathways within the brain. Then, by comparing the homological features of structural and functional brain networks across a large age span, I will highlight the presence of dynamically coordinated compensation mechanisms, suggesting that functional topology is conserved over the depleting structural substrate, and test this conjecture on data coming from a set of different altered brain states (LSD, psylocybin, sleep).
Oct 23 2017 16:00	Room 132	MSc. Lisa Weiss (University of Wien)	Designing a topological filter: Transport of linear and ring polymers in micro-fluidic devices
ABSTRACT: Ring polymers are an important class of biological and synthetic macromolecules [1, 2]. Due to the lack of free ends, they are expected to show distinct behaviour compared to their linear counterparts, as for example with respect to migration, rheology or disentanglement [3]. This simulation study aims at addressing the question whether linear and unknotted ring polymers are transported distinctly in micro- fluidic devices. Hydodynamics is taken into account by employing Multi-Particle Collsion Dynamics [4]. Although a bare slit channel is not sufficient to separate them for all investigated rigidities, we propose a filter by decorating the channel walls with attractive spots. We show that the spots can capture the linear chain while allowing the rings to "roll along the tracks" that these spots form. This mechanism holds true, since spots induce a reorientation of ring polymers close to the decorated surface [5, 6]. In doing so, ring polymers show up to an order of magnitude increase in transport compared to linear chains [6]. At the same time, and for intermediate driving pressure gradients along the channel, a crossover regime appears in which the linear chains are transported faster than the rings due to incessant adsorption-desorption processes that are active for the former but not for the latter. Our work demonstrates the possibility to employ micro-fluidic devices in order to achieve separation of topologically distinct states of polymeric macromolecules. References [1] Lasda, E.; Parker, R. RNA 2014, 20, 18291842. [2] McLeish, T. Science 2002, 1740, 20012002. [3] Kapnistos, M.; Lang, M.; Vlassopoulos, D.; Pyckhout-Hintzen, W.; Richter, D.; Cho, D.; Chang, T.; Rubinstein, M. Nature Materials 2008, 7, 9971002. [4] Malevanets, A.; Kapral, R. Journal of Chemical Physics 1999, 110, 86058613. [5] Poier, P.; Egorov, S. A.; Likos, C. N.; Blaak, R. Soft Matter 2016, 12, 79837994. [6] Weiss, L. B.; Nikoubashman, A.; Likos, C. in preparation 2017, xx, xxxx.
Jul 21 2017 14:00	Room 005	Dr. Jan Smrek (Max Planck Institute for Polymer Research)	Small activity differences drive phase separation in mixtures of active and passive polymers
ABSTRACT: Life is inherently out of equilibrium. Here we study nonequilibrium process of phase separation of active and passive polymers that can be responsible for chromatin organization into active and inactive nuclear compartments [1,2]. Recent theoretical studies found that mixtures of active and passive colloidal particles phase separate but only at very high activity ratio [3, 4]. The high value casts doubts on the biological significance of the phenomenon. However, we show using simulations that when the active and passive particles are polymers, the critical activity ratio decreases with the polymer length [5]. This makes the effect relevant for the DNA organization in living cell nuclei: long-enough actively transcribed DNA strands would separate from inactive strands even if they were chemically identical. Besides the biological application, our simple model is interesting for studying the properties of nonequilibrium phase transitions. [1] N. Ganai, S. Sengupta, and G. I. Menon, Nucleic Acids Res. 42, 4145 (2014). [2] T. Cremer, M. Cremer, B. Hübner, H. Strickfaden, D. Smeets, J. Popken, M. Sterr, Y. Markaki, K. Rippe, and C. Cremer, FEBS Lett. 589, 2931 (2015). [3] A. Y. Grosberg and J.-F. Joanny, Phys. Rev. E 92, 032118 (2015). [4] S. N. Weber, C. A. Weber, and E. Frey, Phys. Rev. Lett. 116, 058301 (2016). [5] J. Smrek and K. Kremer, Phys. Rev. Lett. 118, 098002 (2017).
Jun 09 2017 14:00	Room 132	Dr. Daniele Granata (University of Copenhagen)	Computing ion channels: new insights from evolution and simulations
ABSTRACT:
Jun 08 2017 14:00	Room 128	Prof. Matej Praprotnik (University of Ljubljani)	Open Boundary Molecular Dynamics Simulation
ABSTRACT: Open molecular systems exchange mass, momentum, and energy with their surroundings. In this talk, I will present our Open Boundary Molecular Dynamics (OBMD) simulation method that opens up the boundaries of a molecular system and allows for equilibrium MD simulations in the grand-canonical ensemble as well as nonequilibrium fluid flow simulations. The flow is introduced via an external boundary condition while the equations of motion for the bulk remain unaltered. To illustrate the robustness of OBMD, I will present simulation results of star-polymer melts at equilibrium and in sheared flow.
May 19 2017 14:00	Room 132	Dr. Edoardo Sarti (NIH, Washington)	EncoMPASS: an Encyclopedia of Membrane Proteins Analyzed by Structure and Symmetry
ABSTRACT:
Mar 17 2017 14:30	Room 128	Dr. Stefano Corni (Universitá di Padova)	Localized surface plasmons and their interaction with molecules: a first principle perspective
ABSTRACT: First principle methods such as time-dependent density functional theory are gaining ground in the investigation of surface plasmons localized on metallic nanoparticles and non-conventional plasmonic materials such as graphene nanoflakes. They are also being used to study how the optical properties of molecules are affected by the interaction with plasmonic nanosystems (e.g., surface-enhanced spectroscopies). On one hand, first-principle methods can be used to address basic questions such as the microscopic nature of the plasmonic excitations themselves (what distinguish a plasmon from a single-particle excitation or from an exciton?). On the other hand, hybrid models can be devised for molecules close to metal nanostructures (possible in the presence of a solvent), where the molecule is treated by a first principle method and the nanostructure and the solvent as continuous dielectrics. Such models can give access to the real-time optical behavior of the molecule plus plasmonic nanostructure system when interacting with short light pulses. In this seminar I will discuss our proposals to characterize the plasmonic nature of excitations by means of simple indexes that exploit results of first-principle calculations [1]. We also introduce some on-going developments in hybrid models towards a real-time description of the molecule-metal nanoparticle systems [2]. [1] L.Bursi, A. Calzolari, S. Corni, and E. Molinari ACS Photonics 1, 1049-1058 (2014); ACS Photonics 3, 520-525 (2016). [2] S. Pipolo, S. Corni Real-Time Description of the Electronic Dynamics for a Molecule Close to a Plasmonic Nanoparticle J. Phys. Chem C 120, 28774 (2016)
Mar 10 2017 14:30	Room 005	Prof. Henri Orland (CEA, Saclay, France)	Searching for Transition Paths
ABSTRACT:
Feb 24 2017 11:00	room 005	Dr. Alfredo Braunstein (Politecnico di Torino)	Inference and optimization in statics and dynamics
ABSTRACT: The availability of new types of data allow to parametrize with unprecedented detail systems in many areas of science that are inherently heterogeneous and complex. Predicting the behavior and controlling such systems poses new exciting theoretical challenges. In this talk, I will present two families of strongly related inference and optimization problems with natural motivation in data: the first family involves control and inference on discrete dynamics over networks (namely inference in the Susceptible-Infected-Recovered epidemic model, linear threshold model, decycling and optimal percolation), while the second is related to linear estimation (inference of metabolic fluxes and tomography image reconstruction). I will schematically present Statistical Mechanics approaches to both and some results, including some work in progress and open directions.
Feb 13 2017 14:00	room 128	Dr. Toni Giorgino (Istituto di Neuroscienze, IN-CNR, Consiglio Nazionale delle Ricerche, Italy)	Statistical learning and its applications to biomedicine
ABSTRACT: Statistical learning methods encode data extracted from noisy domains (e.g., observations of the real world) into models, thus condensing information and enabling predictions. Data-driven model development (machine learning) is extensively used in the fields that attempt to understand underlying processes, forecast future events, or plan decisions in presence of noisy data. The seminar will provide an overview of several classes of methods rooted in statistics and artificial intelligence, and how they are used in biomedicine; it will present examples of applications in domains as diverse as time series classification, genotype-based risk prediction in oncology, and in-silico protein-drug interaction studies.
Feb 06 2017 11:00	room 005	Dr. Heinrich Huber (Royal College of Surgeons in Ireland, Dublin)	Biophysics and Systems Biology of Programmed Cell Death and Inflammation in Cancer and Cardiovascular Diseases
ABSTRACT: With modern biology becoming a science of molecular information, expertise from biophysics and systems sciences has pervaded modern biomedical research, providing an exciting fields for physicist, chemist, bioinformaticians and engineers. I here will present examples as to how these fields can bring new insights into modern biomedicine, focusing on my work on molecular signal transduction analysis in programmed cell death and inflammation. Illustrations will include how use of ordinary differential equations and cellular automata in combination with quantitative biochemistry and live cell microscopy will provide a way for explaining failure in cytotoxic chemotherapy and may reveal novel ways for diagnostics and therapy in oncology. Using these signal transduction techniques, I will further provide examples of how tight regulation of inflammation in the body can be understood by studying the interplay of antagonizing bio-molecules, giving insights in the pathophysiology of hyperinflammation after infection and myocardial infarct. Employing methods from quantum chemistry, I will extend the idea of processing signals within biomolecules rather than between them, thereby providing a new paradigm for biological signal processing. I will finalize my talk by exemplifying how specialized solutions for automated high speed confocal microscopy can be engineered and employed to discover new molecular effects relevant to programmed cell death and inflammation.
Feb 01 2017 2:00	room 128 (cinema)	Dr. Luca Dall'Asta (Politecnico di Torino)	Network dismantling
ABSTRACT: Network dismantling consists of determining a minimal set of vertices whose removal leaves the network broken into connected components of sub-extensive size. I will show that for a large class of random graphs, this problem is tightly related to the decycling problem (the removal of vertices, leaving an acyclic graph), and that the latter can be effectively and efficiently solved using algorithmic methods from the statistical mechanics of disordered systems. The results reveal that the problem is intrinsically collective and optimal dismantling sets cannot be viewed as a collection of individually well-performing nodes as usually suggested by most heuristic (greedy, centrality-based) methods. Finally, I will discuss in detail the idea behind the algorithmic technique, that can be used to tackle other interesting inference and optimization problems involving epidemic-like dynamical processes on graphs. Reference: A. Braunstein, L. DallAsta, G. Semerjian, L. Zdeborova, PNAS, 113(44) 12368-12373 (2016)
Jan 23 2017 2:00	room 128-129 (cinema)	Dr. Purushottam Dixit (Department of Systems Biology, Columbia University Medical Center - New York, 10032)	Recombination driven genome evolution and population structure of bacterial species
ABSTRACT: Two processes govern diversification of bacterial genomes. While point mutations always increase population diversity, horizontal gene transfers act as a cohesive force at the population level. Currently, a detailed quantitative understanding of how these two opposing forces shape bacterial evolution at the level of individual genes, genomes, and populations is lacking. In a theoretical model, we identify two qualitatively distinct phases in the dynamics of genome evolution, characterized by the second eigenvalue of the Markov process describing evolution. In the divergent phase the cohesion due to recombination is not sufficient to overcome mutational drift. As a consequence both individual genes and the entire genomes within the same species keep diverging from each other in the course of evolution. At the population level, transient clusters of sub-populations are continuously formed and dissolved. In contrast, in the {\it metastable} phase, the recombination has the upper hand. In this phase, genomes of descendants of a pair of sister cells remain close to each other for long periods of time but eventually escape the pull of recombination and diverge indefinitely. The population of the entire species remains genetically cohesive and stable over time and does not fragment into sexually isolated sub-populations. Real bacterial examples are discussed as well.
Jan 17 2017 11:00	room 005	Dr. Juergen Zanghellini (Austrian Centre of Industrial Biotechnology, Austria)	From (metabolic) networks to functions: the convex geometry of metabolic pathways
ABSTRACT: In the last twenty years the development of constraint-based modeling (CBM) approaches contributed tremendously to our understanding of metabolic processes, pathways, and networks and has become one of the most (if not the most) successful modeling approach in systems biology. Key to this success is the analysis of (genome-scale) metabolic reconstructions. Combined with CBM approaches, these models provide a mechanistic basis to investigate and elucidate genotype-phenotype relationships. In particular elementary flux modes (EFMs) emerged as a successful formal concept to describe metabolic pathways in a mathematically unambiguous manner. I will introduce basic mathematical properties of EFMs and show that they fully characterise the metabolic capabilities of an organism. This is extremely useful specifically with respect to the design of rationally engineered optimal cell factories. However, the calculation of EFMs in metabolic models is known for its combinatorial complexity, which makes their calculation intractable in genome-scale networks. Recently, we introduced thermodynamic EFM analysis (tEFMA), which integrates the metabolome into the EFM analysis. With tEFMA thermodynamically feasible EFMs can be calculated reliably and efficiently even in large-scale networks as their number is significantly smaller than the total number of EFMs. I will demonstrate the biological relevance of our approach by correctly identifying infeasible pathways in E. coli and by unambiguously explaining the experimentally observed behaviour of glutamate dehydrogenase under different environmental conditions. Moreover, I will show that only a few out of all thermodynamically feasible EFMs are biologically relevant and can be combined into thermodynamically feasible flux distributions. These largest, thermodynamically consistent sets of EFMs can be identified by linear programing. Furthermore, by considering commonly available phenotypic data, I will show that only a handful of these sets can contain a biologically relevant solution and that the biologically relevant sets are characterised by their ability to maximise biomass and ATP production, consistent with evolutionary interpretations of cell behaviour.
Jan 09 2017 14:00	Room 005	Dr. Francesco Napolitano (Telethon Institute of Genetics and Medicine (TIGEM), Napoli, Italy)	Comparative Transcriptomics: pushing Data Analysis towards explaining, rather than predicting, cellular mechanisms.
ABSTRACT: Omic technologies and dedicated data analysis techniques allowed to study the cell as a complex system of interconnected molecular entities and events, in which seemingly chaotic interactions give rise to a coherent and self-preserving behavior. Microarray data, summarizing the amount of transcript produced from each gene in the genome at a given time point, can be regarded as a readout of such system. While extensive pharmacological perturbations of the transcriptome have been performed with the main aim of searching for new therapeutic small molecules (drugs), prior knowledge about the used perturbagens can conversely be exploited to investigate biological mechanisms driving the observed cellular response. This talk will discuss the concepts above in the general context of Systems Biology, with applications to genetic diseases. Black-, gray-, and white-box Data Analysis / Machine Learning approaches, with their different strengths and weaknesses, will be demonstrated as powerful tools both to search for therapeutic perturbagens and to reverse-engineer the cellular system in terms of interconnected molecular pathways. Black-box approach limitations will be discussed as particularly undesirable. The use of data integration techniques at different levels to either enrich the starting information base or improve its signal-to-noise ratio will also be showed.
Nov 10 2016 11:30	Roomm 132	Dr. Kwanghyok Jong (SISSA)	Hydrogen Bond Networks and Hydrophobic Effects in the Amyloid beta Chain in Water: A Molecular Dynamics Study
ABSTRACT: In this work, we use classical molecular dynamics (MD) simulations to explore the conformational landscape of the Amyloid Beta 30-35 sequence made up of 6 amino acids which is C-terminal hydrophobic peptide of full-length Amyloid Beta. We use well-tempered metadynamics to explore the free energy landscape of this chain in explicit water with particular focus on understanding the importance of salt-bridges between the N and C termini on the conformational landscape of the peptide. We find that the structural disorder in the conformational landscape is stabilized by a diversity of different interactions such as polar interactions between termini, termini and backbone, and backbone and backbone and non-polar interactions between sidechain groups. We also investigate the properties of the hydrogen bond network surrounding the peptide, providing new insights into the coupling of protein and water motions for this system. This analysis shows subtle differences in this network such as the shortening or lengthening of the wires connecting different parts of the peptide during the conformational fluctuations.
Oct 03 2016 14:30	Room 132	Prof. De Witt Sumners (Florida State University, USA)	Conservation of Writhe Helicity Under Anti-Parallel Reconnection
ABSTRACT: Reconnection is a fundamental event in many areas of science, from the interaction of vortices in classical and quantum fluids, and magnetic flux tubes in magnetohydrodynamics and plasma physics, to recombination in polymer physics and DNA biology. By using fundamental results in topological fluid mechanics, the helicity of a flux tube can be calculated in terms of writhe and twist contributions. We prove that the writhe is conserved under anti-parallel reconnection. Hence, for a pair of interacting flux tubes of equal flux, if the twist of the reconnected tube is the sum of the original twists of the interacting tubes, then helicity is conserved during reconnection. Thus, any deviation from helicity conservation is entirely due to the intrinsic twist inserted or deleted locally at the reconnection site. This result has important implications for helicity and energy considerations in various physical contexts. We will discuss the mathematical similarities between reconnection events in biology and physics, and the relationship between iterated reconnection and curve topology. In particular, the minimal reconnection cascade from (2,2k+1) torus knots to (2,2k) torus links to the unlink of two unknotted circles observed in DNA site-specific recombination is also observed in fluid vortex reconnections. We will also ciscuss helicity and reconnection in a tangle of confined vortex circles in a superfliud. Acknowledgements: Financial support from a Simons Foundation Collaboration Grant for Mathematicians is kindly acknowledged. References: [1] Laing C.E., Ricca R.L., & Sumners D.W. (2015) Conservation of writhe helicity under anti-parallel reconnection, Nature Scientific Reports 5 : 9224 \| DOI: 10.1038/srep09224.
Sep 26 2016 14:00	Room 132	Dott. Maria Sara Bernardi (University of Milan)	Spatial regression models with partial differential operator penalization
ABSTRACT: We present a class of models for surface estimation from noisy data. The models are based on the idea of regression with partial differential regularization. The numerical implementation exploits advanced computing techniques, specifically it uses the Finite Element method. Among the various modeling features, the method is able to deal with spatial domains featuring strong concavities, interior holes and other complex geometries. Moreover, prior knowledge on the phenomenon under study can be included in the model. In this talk, we focus on two extensions of the model. First, we extend the model to accurately estimate surfaces which show spatial anisotropy. We manage to estimate the direction and intensity of anisotropy. Hence, we use the estimated parameters in the partial differential operator of the roughness penalty to obtain a precise estimate of the spatial field. Secondly, we extend the model for the analysis of spatio-temporal data, such as spatially dependent curves or time dependent surfaces.
Sep 23 2016 11:45	Room 132	Prof. Hue-Sun Chan (University of Toronto, Canada)	Impact of Hydrophobicity on Nonnative Interactions, Evolutionary Switches, and Effects of Hydrostatic Pressure in Protein Folding
ABSTRACT: Hydrophobic interactions are a major -- though not necessarily the dominant -- driving force in protein folding. I will present recent advances in applying atomic and coarse-grained simulation approaches to investigate the impact of hydrophobicity on protein folding, using primarily three systems to gain insight into the underlying general principles: (i) Bacterial colicin-immunity proteins Im7 and Im9 are homologous, but they fold by different mechanisms. Im7 tends to fold in a three-state manner via an intermediate but Im9 folding is two-state-like. Our model suggests that nonnative effects in Im7 folding are caused by a higher local hydrophobicity concomitant with a lower local native contact density. (ii) How novel folds of proteins may evolve is addressed by modeling the folding behaviors of 12 experimentally well-characterized GA/GB sequences covering a switch from an all-alpha GA fold to an entirely different four-beta + alpha GB fold. In agreement with experiment, our model exhibits conformational switching upon a single L45Y substitution. The fold preference shows a gradual sequence-dependent change in our model, in line with the latent evolutionary potential concept [2]. Our theoretical perspective thus provide a coherent physical picture for rationalizing and predicting nonnative effects and conformational switches. (iii) Based on explicit-water atomic simulation of volumetric effects of hydrophobic association, an implicit-water atomic chain model is developed to embody effects of the particulate nature of water and other aspects of solvation on a polypeptide. The model was tested by Langevin dynamics simulations of a 16-residue polyalanine. Consistent with prior experiment, coil-helix transition in our simulation is associated with an average volume decrease; but the transition process entails a robust positive activation volume. Thus pressure likely stabilizes helices of short peptides but is expected to slow down their formation. General ramifications of our theoretical findings for pressure effects on globular protein folding will also be explored.
Jul 21 2016 17:30	Room 5	Massimiliano Bonomi (University of Cambridge, UK)	Tackling sampling and accuracy issues in biomolecular simulations with Metadynamic Metainference
ABSTRACT:
Jul 20 2016 14:00	Room 132	Doga Findik (Bogazici University, Istanbul, Turkey)	Ligand effect on enzyme collective dynamics
ABSTRACT: The effect of ligand binding on global vibrational modes were investigated based on the enzyme triosphosphate isomerase (TIM). For this aim, mixed coarse-grained elastic network model ENM (MCG_ENM) was used to observe the frequency shifts in the most collective modes due to the presence of the ligand. First analysis was performed on previous blind docking study, in which various ligands docked on three different TIM conformers. Several different binding sites/poses were detected including the tunnel region, catalytic site and others. The inhibitors bound to the previously identified tunnel region could be differentiated based on their impact on global vibrational modes of the enzyme. Later, six independent MD runs were analyzed by performing MCG_ENM on a total of 54,000 MD snapshots of TIM. Through this computationally efficient technique, altered collective modes and positive shifts in eigenvalues were detected in the complex runs due to the constraining effect of inhibitor binding at the tunnel region. Lastly, a new computational technique was introduced for scanning protein side chains in terms of their constraining effect on ENM modes. In this methodology, the ligand binding effect in the vicinity of a specific residue is mimicked by adding extra nodes to its side chain atoms. ENM-based scanning of TIM also pinpointed to the tunnel region as a key binding site that can alter global dynamics of the enzyme. Scanning of other enzymes indicated marked constraining effect around the ligand positions observed in crystal structures.
Jul 18 2016 14:00	Room 5	Massimiliano Di Ventra (Department of Physics, University of California San Diego, USA)	Memcomputing: a brain-inspired topological computing paradigm
ABSTRACT: Which features make the brain such a powerful and energy-efficient computing machine? Can we reproduce them in the solid state, and if so, what type of computing paradigm would we obtain? I will show that a machine that uses memory to both process and store information, like our brain, and is endowed with intrinsic parallelism and information overhead - namely takes advantage, via its collective state, of the network topology related to the problem - has a computational power far beyond our standard digital computers [1]. We have named this novel computing paradigm memcomputing [2]. As an example, I will show the polynomial-time solution of prime factorization and the NP-hard version of the subset-sum problem using polynomial resources and self-organizable logic gates, namely gates that self-organize to satisfy their logical proposition [3]. I will also show that these machines are described by a Witten-type topological field theory and they compute via an instantonic phase where a transient long-range order develops due to the effective breakdown of topological supersymmetry [4]. The digital memcomputing machines that we propose are scalable and can be easily realized with available nanotechnology components. [1] F. L. Traversa and M. Di Ventra, Universal Memcomputing Machines, IEEE Transactions on Neural Networks and Learning Systems, 26, 2702 (2015). [2] M. Di Ventra and Y.V. Pershin, Computing: the Parallel Approach, Nature Physics, 9, 200 (2013). [3] F. L. Traversa and M. Di Ventra, Polynomial-time solution of prime factorization and NP-hard problems with digital memcomputing machines, arXiv:1512.05064. [4] M. Di Ventra, F. L. Traversa and I.V. Ovchinnikov, Topological field theory and computing with instantons (in preparation).
Jul 08 2016 11:00	Room 132	Prof. Chiara Cappelli (Scuola Normale Superiore di Pisa)	Integrated QM/polarizable MM/Continuum approaches to Molecular Properties and Spectroscopies of Strongly Interacting Solute-Solvent Systems
ABSTRACT:
Jun 10 2016 15:00	Room 5	Prof. Artem Oganov (Center for Materials by Design, State University of New York at Stony Brook)	Computational materials discovery with evolutionary algorithm
ABSTRACT:
May 25 2016 12:30	Room 128-129	Dr. Jakub Otwinowski (University of Pennsylvania)	Constraints on the rate of evolution
ABSTRACT:
May 06 2016 14:00	Room 005	Prof. Martin Lenz (Univ. Paris-Sud, Orsay)	Disordered actomyosin contracts in unexpected ways
ABSTRACT: The motion of living cells is in large part due to the interaction of semi-flexible actin filaments (F-actin) and myosin molecular motors, which induce the relative sliding of F-actin. It is often assumed that this simple sliding is sufficient to account for all actomyosin-based motion. While this is correct in our highly organized striated muscle, we question the application of this dogma to less ordered actomyosin systems, thus reexamining a cornerstone of our understanding of cellular motion.
May 02 2016 14:30	Room 132	Dr. Daniela Selvi (University of Florence)	Automatic systems for epileptic seizure prediction
ABSTRACT:
Apr 28 2016 12:00	Room 005	Dr. Antonio Scialdone (EMBL-EBI, Cambridge, UK)	Using single cell approaches to understand cell fate decisions in early embryo development
ABSTRACT: In most early metazoan embryos, progenitor cells differentiate into precursor cells of major organ systems (blood, heart, etc.) through a process called gastrulation. During gastrulation, pluripotent cells migrate through the primitive streak in a complex spatio-temporal pattern to form the three germ layers (mesoderm, endoderm and ectoderm) that lay the foundation of the basic body plan. Given the inaccessibility and the extremely limiting number of cells in the embryo, traditional experimental approaches for transcriptome analysis cannot be applied to study the diversification of cells during early gastrulation directly in the embryo. Hence, the attention has been focussed on in vitro systems that, however, cannot recapitulate the dynamic of developmental processes in vivo. We used single-cell RNA-sequencing (scRNA-seq) to investigate mesodermal lineage development from hundreds of single cell transcriptomes, covering a developmental time-course from early gastrulation to the generation of primitive red blood cells. I will discuss how, for the first time, we characterised mesoderm formation in vivo with the identification and the description of multiple new cell subpopulations. We computationally reconstructed the spatial locations of cells along the primitive streak and the developmental trajectory leading to embryonic blood formation, in the first in vivo study of primitive erythropoiesis at the single-cell level. Moreover, I will show how scRNA-seq data can be used in combination with knockout experiments to get sharper insights into the cell fate decisions.
Apr 20 2016 14:00	Room 132	Dr. Vittore Scolari (Institut Pasteur, Paris, France)	The interplay amongst xenogene silencing and chromosome architecture in E. coli
ABSTRACT:
Apr 18 2016 14:00	Room 132	Dr. Pilar Cossio (MPI Frankfurt)	Effect of the pulling device in single-molecule force spectroscopy experiment
ABSTRACT: In typical force spectroscopy experiments, a small biomolecule is attached to a soft polymer linker that is pulled with a relatively large bead or cantilever. At constant force, the total extension stochastically changes between several values, indicating that the biomolecule undergoes transitions between conformational states. Here, We consider the influence of the dynamics of the linker and mesoscopic pulling device on the force-dependent rate of the conformational transition extracted from the time-dependence of the total extension. We derive analytic expressions for the observables that account for the mechanical response and dynamics of the pulling device and linker. Possible artifacts arise when the characteristic times of the pulling device and linker become comparable to, or slower than, the lifetimes of the metastable conformational states, and when the highly anharmonic regime of stretched linkers is probed at high forces. The theory provides a framework for both the design and the quantitative analysis of force spectroscopy experiments by highlighting, and correcting for, factors that complicate their interpretation.
Dec 17 2015 14:00	Room 132	Dr. Giulia Palermo (EPFL Lausanne, CH)	A Ruthenium-Arene Compound Displays a Novel Mode of One-Stranded Intercalation that is DNA Topology-Dependent
ABSTRACT: Much recent work on developing improved chemotherapeutic compounds has focused on new ruthenium-based agents, which may provide more effective alternatives to platinum drugs. Although ruthenium agents are typically much less cytotoxic than platinum drugs, [(η6-THA)Ru(ethylenediamine)Cl][PF6] (THA= 5,8,9,10-tetrahydroanthracene; RAED-THA-Cl[PF6]), is remarkable in displaying a high antiproliferative activity similar to cisplatin, in spite of having only monofunctional DNA-binding capability. This has been attributed to intercalation of the THA group in adducts formed on naked DNA. However, using crystallographic methods and molecular dynamics simulations, we show here that the THA group in fact remains solvent exposed and does not disrupt base stacking in RAED-THA adducts on naked DNA, whereas adducts formed in the nucleosome core are associated with a novel, one-stranded intercalation and DNA distortion mode. The unique RAED-THA adduct structures and their dramatic distinctions between different topological states of DNA may contribute to the unusually high cytotoxicity of this anticancer agent.
Nov 27 2015 14:30	Room 132	Pemra Doruker (Bogazici University, Istanbul, Turkey)	Effect of ligand binding on enzyme vibrational dynamics
ABSTRACT: The mixed resolution elastic network model, where the ligand is at atomistic detail and the protein is coarse-grained, will be presented in this talk as an efficient approach for assessing the ligand binding effect on enzyme dynamics (Kurkcuoglu et al. Biophys J., 2009, 2015). For the specific case of triosephosphate isomerase (TIM), collective modes dictated by the dimeric enzymes topology have been shown to guide its functional loop dynamics, which is critical for substrate entrance, product release and catalysis. Application of mixed resolution ENM on more than 50,000 TIM conformers obtained from independent molecular dynamics runs and blind docking studies indicates that certain inhibitors (benzothiazoles) bound to the interfacial tunnel region alter enzymes collective modes and cause positive shifts in eigenvalues. Such a constraining effect seems coupled to the enhancement of specific interactions at TIMs interface and allosteric changes observed at its catalytic site including the functional loop. An alternative ENM-based residue scanning method also points to the tunnel as a key region for altering the global modes of TIM. Thus, the extent to which different type of ligands, including allosteric and orthosteric ones, modify the low-frequency vibrational modes of enzymes can be quantified by mixed resolution ENM. Such an approach may be further utilized for isolating among alternative ligand binding positions/poses those that have an effect on enzyme global dynamics.
Nov 25 2015 10:00	Room 133	Nils Hildebrand (Hybrid Materials Interfaces Group - Faculty of Production Engineering - University of Bremen)	Protein Adsorption on Oxides using Classical Molecular Dynamic
ABSTRACT: The physisorption of chymotrypsin on amorphous silica is investigated by classical Molecular Dynamics (MD) methods in comparison to adsorption and circular dichroism (CD) experiments. The long-range protein-surface attraction field is calculated in an implicit solvent based on DLVO theory. These calculations reveal a preferred protein orientation, which could be confirmed in explicit solvent simulations. Driven by its large dipole moment, chymotrypsin adsorbs with its alpha-helical regions pointing towards the surface. Positively charged hydrophilic residues form dominant binding motifs by adsorbing in dense water layers around the deprotonated silanol surface groups. No significant conformational changes are observed in MD simulations lasting 300 ns. In order to capture surface-induced conformational changes revealed by CD experiments, parallel tempering in combination with metadynamics is employed. In these simulations, the helical content of chymotrypsin is used as a reaction coordinate.
Nov 11 2015 12:00	Room 132	Dr. Francesco Ferrari (IFOM, Milan)	Chromatin architecture and transcription dynamics
ABSTRACT: All cells within a multicellular organism share an almost identical genome, but they differentiate to have distinct functions and phenotypes. Cell identity is defined by the control of genome functionality and in particular by epigenetic and transcriptional regulation. The 3D organization of chromatin within the cell nucleus is also crucial to define genome functionality. Our knowledge of the role of chromatin architecture has greatly advanced over the last decade with the development of chromatin conformation capture (3C) and its high-throughput derivatives (such as 4C, 5C and HiC). Other experimental techniques based on high-throughput sequencing have similarly allowed great leaps in understanding transcription and epigenetic regulation on a genome-wide scale. We will discuss recent work from our lab and in particular how the development of specific computational biology solutions has been instrumental to lead advancements in these fields.
Jul 15 2015 14:00	Room 005	Dr. Raffaello Potestio (Max Planck Institute for Polymer Research)	Computer simulations of soft matter: bridging the scales
ABSTRACT: A feature common to most soft matter systems, ranging from simple liquids to large biomolecules, is the interplay of characteristic length and time scales, which determines their mechanical and dynamical properties. This multi-scale nature limits our capability to study them by means of computer simulations: in fact, the size of the system often makes it impossible to treat the whole of it at the full-atom level; at the same time, coarse-grained models might lack relevant chemical details. In this talk I will discuss some of the strategies available in the computational study of soft matter. Particular emphasis will be given to adaptive resolution techniques, which allow the concurrent use of different models of the system at different levels of resolution.
Jul 13 2015 16:00	Big Meeting Room, 7th floor	Prof. Erel Levine (Harvard University)	Interactions and Complexity of Small RNA
ABSTRACT: Over the last years it had become clear that small RNA molecules play a central role in biology that has been overlooked for decades. Small RNA systems are characterized by high density of interactions at multiple scales, that collectively give rise to function. In this talk I will describe our efforts to to learn how small RNA regulatory functions stem from the structure of these underlying interactions. After a brief review of the relevant biology, I will first address the question how intra-molecular interactions determine the functional properties of a small RNA, using a combination of quantitative large-scale experiment and data-driven modeling. I will then turn to discuss how the evolutionary imprint of genetic interactions can be used to uncover regulatory strategies by solving an inverse problem of statistical mechanics.
Jul 08 2015 14:00	Room 005	Dr. Rute da Fonseca (University of Copenhagen)	DNA from ancient maize reveals early targets of selection during domestication and diffusion routes out of Mexico
ABSTRACT:
Jun 23 2015 11:00	room 132	Dr. Pavel Baná (Department of Physical Chemistry, Palacky University Olomouc, Czech Republic)	RNA tetra loops - splendours and miseries of MD simulations
ABSTRACT:
May 18 2015 14:00	Room 132	Dr. Sabine Reißer (Institute of Physical Chemistry, Theoretical Chemical Biology, Karlsruhe Institute of Technology (KIT))	Computational studies on membrane-active antimicrobial peptides and comparison with NMR data
ABSTRACT: Antimicrobial peptides are promising candidates to solve the problem of multiresistant bacteria, since bacteria can develop resistances against them much worse than against conventional antibiotics. The reason is that these short (<= 40 amino acids) amphiphilic molecules directly disturb the bacterial cell membrane, and don't have a very specific target inside the cell, like conventional antibiotics do. However, antimicrobial peptides can potentially cause cell lysis also in erythrocytes, and some can even enhance the formation of biofilms. It is therefore crucial to understand the peptide-membrane interactions to create or find antimicrobial peptides which lack these side effects. The interactions between membranes and peptides can be studied by solid state NMR in oriented lipid bilayers. In our lab, we use quadrupolar couplings from 2H NMR and dipolar couplings from 19F NMR to elucidate the peptide structure and orientation in the membrane at different peptide concentrations, temperatures and in different lipids. I use computational methods to reproduce these experimental data and to help the interpretation on the atomistic level. With long MD simulations in lipid membranes, I calculated structural parameters describing the peptide structure, which are essential for the evaluation of NMR data. For the prediction of the membrane-inserted orientation of peptides, I developed the hydrophobic moment vector method, which uses the electrostatic potential on the molecular surface to calculate a vector which points towards the most hydrophobic part of the peptide and which aligns with the membrane normal (1). I used this method also to explain the structure-function-relationship of photo-switchable analogs of the cyclic antimicrobial peptide gramicidin S. The antimicrobial activity of these analogs can be switched on or off by visible/UV light, offering a possible solution to avoid the aforementioned side effects (2). (1) Reißer, S., Strandberg, E., Steinbrecher, T., Ulrich, A. S. (2014): 3D Hydrophobic Moment Vectors as a Tool to Characterize the Surface Polarity of Amphiphilic Peptides. Biophys. J. 106: 23852394. (2) Babii, O., Afonin, S., Berditsch, M., Reißer, S., Mykhailiuk, P. K., Kubyshkin, V. S., Steinbrecher, T., Ulrich, A. S., Komarov, I. V. (2014): Controlling Biological Activity with Light: Diarylethene-Containing Cyclic Peptidomimetics. Angew. Chem. Int. Ed. Engl. 53(13): 33923395
May 13 2015 14:00	Room 132	Petr Stadlbauer (Institute of Biophysics, Academy of Sciences of the Czech Republic, Brno, Czech Republic)	Folding of guanine quadruplexes studied by molecular dynamics simulations
ABSTRACT: Guanine quadruplexes are non-canonical four-stranded DNA structures that seem to be widespread in the human genome, most notably in promoter sequences of genes and at chromosome ends in the telomeric region. While diverse quadruplex topologies have been well characterized experimentally, less is known about their folding kinetics. Several folding pathways have been proposed based on kinetics measurements, but none of the experimental methods provide satisfactory resolution of the suggested intermediate states. We have used extended set of molecular dynamics simulations to complement the folding picture by providing atomistic insight into the formation of human telomeric quadruplexes and some other quadruplex structures as well. We have observed that parallel stranded quadruplexes likely undergo formation via mutual strand slippage of its strands with no distinct intermediates, whilst folding of antiparallel quadruplexes involve various triplex and hairpin intermediates. Nevertheless, our data suggest that myriads of other misfolded structures participate in the folding process, which were not taken into account in the previously proposed folding pathways.
May 07 2015 11:00	Room 004	Prof. Peter Bolhuis (University of Amsterdam)	Free energy, kinetics and mechanisms of protein self-assembly.
ABSTRACT:
May 05 2015 14:00	Room 005	Prof. Peter Bolhuis (University of Amsterdam)	Sampling rare event pathways: from single barriers to kinetic networks.
ABSTRACT:
Apr 17 2015 14:00	Room 005	Dr. Mattia Zampieri (ETH Zurich, Switzerland)	Metabolomics reveals the multilevel response to antibiotic perturbations
ABSTRACT: Microbes have shown a remarkable ability in evolving mechanisms to evade the killing actions of antimicrobial agents, such that treatment of bacterial infections represents again an urgent global challenge. By combining a new mass-spectrometry platform for high-throughput metabolomics with novel computational frameworks, we reveal the active role of metabolism in mediating the immediate bacterial stress response to antimicrobials, and on a long term, its crucial role in shaping evolutionary trajectories for antibiotic resistance. Integrating dynamic metabolic profiling of the E.coli response to antibiotic perturbations with a large compendium of static metabolome data of E.coli gene deletion mutants, we inferred potential new and less conventional targets to interfere with metabolic subsystems involved in the common adaptive response to antimicrobials.
Apr 16 2015 14:00	Room 005	Marco Garavelli (Ecole Normale Superieure Lyon & Universita' degli Studi di Bologna)	Modelling light induced events in complex environments: moving towards an accurate computational photochemistry, photobiology and spectroscopy
ABSTRACT: The use of the computer to simulate light induced events in photoactive molecular materials has given access to a detailed description of the molecular motions and mechanisms underlying the reactivity of organic and bio-organic chromophores. Thus, different computational strategies and tools can now be operated like a virtual spectrometer to characterize and understand the photoinduced molecular deformation and reactivity of a given dye, allowing for a rational of the recorded time-resolved observations and an accurate description of photochemical/photobiological processes. This contribution reviews our recent advances in this field, by presenting our studies in modeling the tuning/controlling effects of the environment (i.e. the solvent, the protein, the surrounding structure in general) on the photophysical/photochemical properties of organic chromophores and biological photoreceptors. The paradigmatic case of retinal systems and visual proteins [1-2] will be presented among the others. The info collected above is then exploited for the design of novel photoactive molecular and soft materials, including a novel paradigm of electrochromism for applications in a new generation of color tunable displays and e-ink devices. Additionally, here we report our recent methodological advances in the modeling of sized photoactive systems by illustrating a new implementation of a general hybrid approach with a modular structure that is able to integrate some specialized softwares and acts as a flexible computational environment, eventually allowing for so far inaccessible calculations (e.g., non-adiabatic molecular dynamics of unprecedented accuracy on large molecular materials). Finally, our latest achievements in developing (and modeling) non-linear bi-dimensional electronic spectroscopy as a novel diagnostic tools for tracking structural/dynamical problems in complex environments (e.g., biologically relevant systems such as proteins and DNA) will be presented [4-5]. [1] D. Polli, et al. Nature 467 (2010) 440. [2] O. Weingart, et al. PCCP 13 (2011) 3645. [4] I. Rivalta, et al. PCCP 16 (2014) 16865. [5] I. Rivalta, et al. JPCB 118 (2014) 8396.
Apr 13 2015 16:00	Room 132	Dr. Asja Jelic (ICTP)	Superfluid transfer of information in starling flocks
ABSTRACT:
Apr 01 2015 14:00	Room 005	Matthias Maier (University of Heidelberg, Germany )	Model adaptation in context of multiscale methods
ABSTRACT: A large class of modeling problems in Physics and Engineering is of multiscale character, meaning, that relevant physical processes act on highly different length scales. This usually implies high computational cost for a full resolution of the problem. One way to avoid such a full resolution are multiscale schemes, where, generally speaking, an effective model is solved on a coarse scale with upscaled, effective parameters. Those paramaters are determined with the help of localized sampling problems on a fine scale. Multiscale schemes introduce significant complexity with respect to sources of error, not only are there discretization errors on a coarse and fine scale, but also a model error introduced by the modeling assumption. This makes suitable a posteriori strategies highly necessary. In this talk different model adaptation strategies for the Variational Multiscale Method (VMM) and the Heterogeneous Multiscale Method (HMM) are examined and a general framework for model adaptation (based on the HMM) is introduced. The framework is derived within the setting of ``goal-oriented'' adaptivity given by the so-called Dual Weighted Residual (DWR) method. Based on the framework a sampling-adaptation strategy is proposed that allows for simultaneous control of discretization and model errors with the help of classical refinement strategies for mesh and sampling regions. Further, a model-adaptation approach is derived that interprets model adaptivity as a minimization problem of a local model-error indicator.
Mar 25 2015 14:00	Room 132	Farsheed Rafiee (Department of Physics, Institute for Advanced Studies in Basic Sciences (IASBS), Zanjan, Iran )	Quantitative Model for Nucleosome Positioning and Dynamics
ABSTRACT: Three-quarters of eukaryotic DNA are wrapped around protein cylinders forming so-called nucleosomes that block the access to the genetic information. Nucleosomes need therefore to be repositioned, either passively (by thermal fluctuations) or actively (by molecular motors). In addition to the nucleosome repositioning, unwrapping of nucleosome is also possible within the chromatin complex. In this talk, after introducing sequence dependent elastic model, I will talk about the energy of nucleosomal DNA for different DNA sequences using the theoretically determined structure and energy of nucleosomal DNA [D. Norouzi and F. Mohammad-Rafiee, J. Biomol. Struct. Dyn. 32, 104 (2014)]. After that I will emphasize on the dynamical behavior of a mono-nucleosome using a theoretical model that takes into account the nucleosome sliding and the effect of the DNA sequences. Using a dynamical Monte-Carlo simulation algorithm, we show that for an appropriate set of parameters, the results of the model are in quantitative agreement with data from in vitro positioning experiments.
Feb 23 2015 14:00	Big meeting room 7th floor	Dr. Marco Punta (CNRS, Paris)	Function annotation transfer, intrinsic disorder and protein-protein interactions: the importance of family ties in the protein universe
ABSTRACT:
Feb 20 2015 12:00	Room 005	Dr. Vincenzo Carnevale (Temple University, Philadelphia, USA)	Activation of Voltage-Gated-Like Ion Channels: What Can We Learn From Evolution?
ABSTRACT:
Feb 19 2015 11:30	Room 132	Prof. Matthew Turner (Warwick University, UK)	Swarms: connecting animal and thermodynamic systems
ABSTRACT:
Jan 21 2015 16:00	Room 134	Cecilia Laschi (The BioRobotics Institute, Scuola Superiore Sant'Anna, Italy)	Soft Robotics: a new paradigm for robotics and a new challenge for many other disciplines
ABSTRACT: Soft robotics, intended as the use of soft materials in robotics, is a young yet promising and growing research field. The need for soft robots emerged in robotics, for facing unstructured environments, and in artificial intelligence, too, for implementing the embodied intelligence, or morphological computation, paradigm, which attributes a stronger role to the body and its interaction with the environment. Using soft materials for building robots poses new technological challenges: the technologies for actuating soft materials, for embedding sensors into soft robotparts, for modelling and for controlling soft robots are among the main ones. Though still in its early stages of development, soft robotics is finding its way in a variety of applications, where safe contact is a mainissue, in the biomedical field, as well as in exploration tasks and in the manufacturing industry. Literature in soft robotics is increasingly rich, though scattered in many disciplines. The soft robotics community is growing worldwide and initiatives are being taken, at international level, for consolidating this community and strengthen its potential for disruptive innovation.
Dec 17 2014 9:00	room 005	Simone Melchionna (IPCF - CNR, Italy)	From fine grained to coarsed grained models for DNA plasmids
ABSTRACT: Work is in progress to interpret experiments on depositions of DNA plasmids. In order to overcome the large relaxation times involved, a two-level graining procedure allows exploring the bioloplymer structure and, from preliminary data, provides insight into supercoiling at neutral, positive and negative linking numbers.
Dec 17 2014 9:45	room 005	Barbara Capone (University of Vienna)	Multi-blob coarse graining for ring polymer solutions
ABSTRACT: In this talk I will present a multi-scale molecular modeling of concentrated solutions of unknotted and non-concatenated ring polymers in good solvent conditions. The approach is based on a multi-blob representation of each ring polymer, which is capable of over- coming the shortcomings of single-blob approaches that lose their validity at concentrations exceeding the overlap density of the solution.
Dec 17 2014 11:00	room 005	Flavio Romano (University of Oxford)	A coarse-grained model for DNA: properties and applications
ABSTRACT: DNA is a very important molecule in biology due to its fundamental role as genetic storage. Recently, with advances in the possibilities of manipulating its chemistry, DNA has been increasingly used as a material in nanotechnology, both to build structures with a target function or shape as well as active devices. In this contribution I will discuss a coarse-grained model at the nucleotide level which is able to describe both single- and double-stranded DNA at high salt concentrations. Some applications of the model that demonstrate its predicting power will be discussed, such as its force-extension properties, the ability to reproduce topological effects and the extreme bending properties, as well as dynamical properties such as displacemente rates. Finally, I will describe the most recent extensions of the model to lower salt concentrations (including builogical settings) and RNA.
Dec 17 2014 11:45	room 005	Lorenzo Rovigatti (University of Vienna)	Tetravalent DNA Constructs as Valence-limited Soft Building Blocks
ABSTRACT: We investigate, by employing a realistic DNA coarse-grained model and molecular dynamics simulations, the phase behaviour, the structure and the dynamics of DNA tetramers, i.e. tetravalent nanoconstructs entirely made of DNA. We find that, as the system is cooled down, tetramers undergo a gasliquid phase separation in a region of concentrations which, if the difference in salt concentration is taken into account, is comparable with the experimental phase diagram recently measured[1]. By analysing simulation results, we show that the structure and flexibility of the single tetramer is retained for all the investigated temperatures and concentrations. By increasing the concentration, the system remains in a single-phase state and DNA tetramers start to form a tetrahedral network which, at low temperatures, becomes fully bonded. In line with a recent study of the thermodynamics of a tetravalent patchy particle model[3], we provide numerical strong evidence that the network liquid is more stable than crystallines states in a very wide range of concentrations[4]. [1] S. Biffi, R. Cerbino, F. Bomboi, E. M. Paraboschi, R. Asselta, F. Sciortino and T. Bellini, Proc. Nat. Acad. Sci. (2013) [2] L. Rovigatti, F. Bomboi and F. Sciortino, J. Chem. Phys. (2014) [3] F. Smallenburg and F. Sciortino, Nat. Phys. (2013) [4] L. Rovigatti, F. Smallenburg, F. Romano and F. Sciortino, ACS Nano (2014)
Dec 17 2014 14:00	room 005	Peter Cimermancic (UCSF)	Can architectures of macromolecular assemblies be determined by measuring cell colony sizes?
ABSTRACT: To understand the workings of a living cell, we need to know the structures of its macromolecular assemblies. Determining these structures has required pure samples of the studied assembly. I will present an alternative strategy based on in vivo measurements of genetic interactions between the assembly proteins. We show that genetic interactions can be sufficient to define the molecular architecture of an assembly, and are thus comparable in their utility to chemical cross-links.
Dec 17 2014 14:45	room 005	Marco Punta (Université Pierre et Marie Curie, Paris, France)	The Dark Matter of the protein sequence universe: challenges and perspectives
ABSTRACT: The ever increasing pace at which genomes and metagenomes are being sequenced is generating unprecedented amounts of raw biological sequences. Experimental characterisation of functional elements including proteins, functional RNAs and conserved non-coding DNA elements, however, remains an expensive and in most cases painstakingly slow process. While computational analysis plays a pivotal role in generating functional hypothesis for experimentally uncharacterised sequences, to date a significant fraction of known sequences remains without any annotation, be it experimental or computational. In this talk, I will focus on the protein sequence space and discuss some of the challenges that we face in annotating its 'dark matter', including: remote homology detection, domain discovery, pathway annotation and handling of intrinsically disordered regions.
Dec 17 2014 16:00	room 005	Giulia Palermo (École polytechnique fédérale de Lausanne)	Anandamide Hydrolysis in FAAH Reveals a Dual Strategy for Efficient Enzyme-assisted Amide Bond Cleavage via Nitrogen Inversion
ABSTRACT: The fatty acid amide hydrolase (FAAH) is an integral membrane protein responsible for the deactivating hydrolysis of a family of endogenous fatty acid amides, thus interfering with many physiological processes. In recent years, several drug discovery programs have targeted FAAH, as its inhibition has been shown to be effective for the treatment of pain, inflammation and cancer. We combined classical molecular dynamics (MD) and quantum mechanical/molecular mechanics (QM/MM) simulations to unravel the whole catalytic cycle of FAAH in complex with anandamide, the main neurotransmitters involved in the control of pain. While microsecond MD simulations of FAAH in a realistic membrane/water environment provided a solid model for the reactant state of the enzymatic complex, QM/MM simulations depict a highly concerted two-step catalytic mechanism characterized by: (1) acyl-enzyme formation after hydrolysis of the substrate amide bond and (2) deacylation reaction with restoration of the catalytic machinery. We found that a crucial event for anandamide hydrolysis is the inversion of the reactive nitrogen of the scissile amide bond, which occurs during the acylation rate-limiting step. We show that FAAH uses an exquisite catalytic strategy to induce amide bond distortion, reactive nitrogen inversion, and amide bond hydrolysis, promoting catalysis to completion. This new strategy is likely to be of general applicability to other amidases/peptidases that show similar catalytic site architectures, providing crucial insights for de-novo enzyme design or drug discovery efforts.
Dec 17 2014 16:45	room 005	Vincenzo Carnevale (ICMS - Temple University)	Activation and Modulation of Voltage-Gated-Like Ion Channels
ABSTRACT: Ion channels are ubiquitous transducers of chemical and electrical stimuli in cells. In voltage-gated-like ion channels, conformational changes in a domain responsible for sensing a stimulus (transducer) affect the state of a gate domain (effector) that open and close, controlling the flux of permeant ions. The details of the conformational change and allosteric communication to the gate are of great relevance: pharmacological modulators of ion channels such as anesthetics interfere with the coupling between the transducer and effector domains. A molecular-level description of these processes is potentially transformative, as it would enable the design of novel precise and selective allosteric drugs. Motivated by fundamental questions concerning the mechanism of general anesthesia, I will discuss recent all atom MD investigations of the activation and opening processes in voltage-gated-like ion channels. I will also describe some attempts at disentangling the intricate circuitry of energetic and statistical couplings among a functioning channels constituent amino acids in order to highlight the network of residue-residue interactions sustaining the mechanical coupling between distinct protein domains.
Dec 01 2014 14:00	Room A-132	Dr. Hao Wu (Free University of Berlin)	TRAM: Markov model based equilibrium analysis of simulations at multiple equilibrium states
ABSTRACT:
Oct 17 2014 9:00	room 005	Pavel Banas (Palacky University Olomouc (Czech Republic))	Force field performance for description of RNA conformational dynamics.
ABSTRACT:
Oct 17 2014 10:10	room 005	Michele Cascella (University of Oslo (Norway))	Lipophilic transporters of the Sec 14-like family: Structure, dynamics, and function explored by computer simulations.
ABSTRACT:
Oct 17 2014 11:00	Room 005	Federico Fogolari (University of Udine (Italy))	Accuracy assessment of implicit solvent model forces for nucleic acids and nucleic acid-protein complexes.
ABSTRACT:
Oct 16 2014 15:00	room 005	Christophe Zimmer (Pasteur Institute, Paris (France))	Dissecting the yeast nucleus by computational imaging and modeling.
ABSTRACT:
Oct 16 2014 16:10	room 005	Carmen Domene (King's College London (United Kingdom))	Studies of ion conduction through cell membranes using free energy methods.
ABSTRACT:
Oct 16 2014 17:00	room 005	Eivind Hovig and Jonas Paulsen (Oslo University Hospital (Norway))	Inferential 3D genomics.
ABSTRACT:
Oct 15 2014 15:00	room 134	Dr. Gregor Kosec (Parallel And Distributed Systems Laboratory Jozef Stefan Institute, Ljubljana)	Local Meshless Numerical Method for Solving Transport Problems
ABSTRACT: In the last decade, a new class of numerical methods, referred to as the meshless methods, is becoming popular as alternative to classical methods, like Finite Difference Method (FDM), Boundary Element Method (BEM) or Finite Element Method (FEM). A local meshless method (LMM) is based entirely on the approximation constructed over a local subset of scattered computational points and does not require any kind of special topological relations between computational points. The LMM can be formulated in a form suitable for straightforward implementation and further upgrade to treat anomalies such as sharp discontinues or other obscure situations often occurring in complex simulations. Besides simple formulation and implementation, the LMM also enables high parallel efficiency since the localization reduces inter-processor communication, which is often a bottleneck of parallel algorithms. We demonstrate the method on the simulation of (i) natural convection in a free fluid and porous media, (ii) the solidification of a binary alloy, and (iii) the dynamics of charge carrier in a semiconductor device. The basic properties of LMM, i.e. convergence, complexity and intrinsic features, are assessed on the diffusion and Burgers' equations. The parallel efficiency of the presented LMM implementation is demonstrated through the speedup measurements on multicore computer architecture and multi GPU architecture.
Oct 06 2014 14:00	room 132	Dr. Anna Battisti (Università Roma 3)	Transient secondary and tertiary structures in tau, an intrinsically disordered protein
ABSTRACT:
Jul 24 2014 14:00	room 132	Dr. Vojtech Mlynsky (Regional Centre of Advanced Technologies and Materials (RCPTM), Department of Physical Chemistry, Faculty of Science, Palacky University, Olomouc, Czech Republic)	Structure, Dynamics and Reaction Mechanism of RNA Enzymes
ABSTRACT: The source of RNA catalytic power is not widely understood and theoretical approaches (especially molecular dynamic (MD) simulations combined with multiscale quantum mechanical/molecular mechanical (QM/MM) calculations) are able to provide structural and mechanistic description of processes in biomacromolecules. We investigated the structural dynamics and reaction mechanism of RNA backbone self-cleavage catalyzed by two small self-cleaving ribozymes, i.e., hairpin and HDV ribozymes. In the hairpin ribozyme, we observed two different reaction mechanism with the overall barrier in agreement with experiments: (i) combined general acid/general base mechanism, where N1-deprotonated guanine 8 (G8-) and protonated adenine 38 (A38H+) participate directly in the reaction as general base and general acid, respectively, and (ii) proton shuttling scenario, where both catalytically essential nucleobases (canonical G8 and protonated A38H+) are not directly involved in the cleavage and the proton is transferred via nonbridging oxygen of the scissile phosphate. Both mechanism are energetically close and might be in competition. In the HDV ribozyme, we followed reaction path, where the hydroxide ion coordinated to the Mg2+ ion activates the 2'-OH nucleophile. We obtained various activation barriers, which are dependent on the specific position and coordination of the Mg2+ ion in the active site. The QM/MM energies indicate significant pKa shift of the nucleophilic 2'-OH group in the active site of HDV ribozyme that contribute to catalysis due to easier activation of such 2'-OH nucleophile.
Jul 21 2014 12:00	room 132	Cedric Vaillant (Laboratoire de Physique, Ecole Normale Superieure de Lyon, France)	Influence of the genomic sequence on the primary structure of chromatin
ABSTRACT: As an important actor in the regulation of nuclear functions, the nucleosomal organization of the 10nm chromatin fiber is the subject of increasing interest. Recent highresoluted mapping of nucleosomes along various genomes ranging from yeast to human, have revealed a patchy nucleosome landscape with alternation of depleted, well positioned and fuzzy regions. For many years, the mechanisms that control nucleosome occupancy along eukaryotic chromosomes and their coupling to transcription and replication processes have been under intense experimental and theoretical investigation. A recurrent question is to what extend the genomic sequence dictates and/or constrains nucleosome positioning and dynamics? In that context we have recently developed a simple thermodynamical model that accounts for both sequence specificity of the histone octamer and for nucleosomenucleosome interactions. As a main issue, our modelling mimics remarkably well in vitro data showing that the sequence signaling that prevails are high energy barriers that locally inhibit nucleosome formation and condition the collective positioning of neighboring nucleosomes according to thermal equilibrium statistical ordering. When comparing to in vivo data, our physical modelling performs as well as models based on statistical learning suggesting that in vivo bulk chromatin is to a large extend controlled by the underlying genomic sequence although it is also subject to finiterange remodelling action of external factors including transcription factors and ATP-dependent chromatin remodellers. On the highly studied S. cerevisiae organism, we discuss the implications of the highlighted "positioning via excluding" mechanism on the structure and function of yeast genes.
Jul 14 2014 14:00	Room 005	Cedric Vaillant (Laboratoire de Physique, Ecole Normale Superieure de Lyon, France)	Modeling epigenome folding: Formation and dynamics of topologically-associated chromatin domains
ABSTRACT: Genomes of eukaryotes are partitioned into domains of functionally distinct chromatin states. These domains are stably inherited across many cell generations and can be remodelled in response to developmental and external cues, hence contributing to the robustness and plasticity of expression patterns and cell phenotypes. Remarkably, recent studies indicate that these one-dimensional epigenomic domains tend to fold into three-dimensional topologically-associated domains forming specialized nuclear chromatin compartments. However, the general mechanisms behind such compartmentalization including the contribution of epigenetic regulation remain unclear. Here, we address the question of the coupling between chromatin folding and epigenome. Using polymer physics, we analyze the properties of a block copolymer model that accounts for local epigenomic information. Considering copolymers build from the epigenomic landscape of Drosophila, we observe a very good agreement with the folding patterns observed in chromosome conformation capture experiments. Moreover, this model provides a physical basis for the existence of multistability in epigenome folding at sub-chromosomal scale. We show how experiments are fully consistent with multistable conformations where topologically-associated domains of the same epigenomic state interact dynamically with each other. Our approach provides a general framework to improve our understanding of chromatin folding during cell-cycle and differentiation and its relation to epigenetics.
Jul 09 2014 14:00	room 005	Prof. Mark E. Tuckerman (Department of Chemistry and Courant Institute of Mathematical Sciences, New York University, NY)	Studies in structure and transport using molecular dynamics
ABSTRACT:
Jun 27 2014 12:00	room 5	Andrea De Martino (Roma)	A quantitative view of microRNA-based regulation, from the ceRNA hypothesis to system-level effects
ABSTRACT: The observation that, through a sequestration mechanism, microRNAs (miRNAs) can act as mediators of effective interactions among their common RNA targets (competing endogenous RNAs, ceRNAs) has brought forward the idea (`ceRNA hypothesis') that RNAs can regulate each other in extended miRNA-mediated `cross-talk' networks, with enormous potential implications for post-transcriptional regulation (PTR). We shall review recent work aimed at placing the ceRNA hypothesis on quantitative ground. In specific, we have studied a mathematical model where the emergence of cross-talk networks can be characterized in detail both statically and dynamically. We find that ceRNA cross-talk is tunable with miRNA levels, can achieve high flexibility and selectivity at stationarity, and may be amplified during transients under optimal conditions, making it an efficient way for a cell to obtain fast up-shifts in the level of a ceRNA when necessary. Large local responses however typically require sizable transfection-like perturbations, in which case the system develops strongly non-linear, threshold behaviour. By constrast, weaker long-range (system-level) effects develop under broad conditions, and known miRNA-ceRNA binding networks appear to be optimally designed for such a `collective' behaviour to arise. Finally, we show that these insights lead to improved inference of miRNA/ceRNA targeting patterns from transcriptome data. Overall, our results indicate that competition for miRNAs may provide an elementary mechanism to achieve both fast responses on short time scales and weaker but system-level regulatory effects over longer time scales.
Jun 13 2014 11:00	room 131	Vladimir Kravtsov (ICTP)	Anderson localization on the Bethe lattice: non-ergodicity of extended states
ABSTRACT:
May 19 2014 11:00	SISSA - room 128-129	Prof. Loren Dean Williams (The Ribo Evo Center and the School of Chemistry and Biochemistry, Georgia Tech, Atlanta GA)	RNA and Protein
ABSTRACT: Nature records history, in the night sky, the rock record, the rings of trees, and especially in biological molecules. We use biological molecules to study the ancient history of life. We study the oldest macromolecules in biological systems, which are assemblies of RNA and protein called ribosomes. The ribosome originated during the origin of life by assembly of proto-RNA and proto-peptide oligomers that formed the ancestral catalytic core, called the Peptidyl Transferase Center. The prokaryotic ribosome was fully mature and fully functional at LUCA, establishing the common core of all ribosomes in extant biology. In Eukaryotes, the surface of the ribosome continued to evolve post- LUCA. In modern Mammalia, the ribosome has reached a zenith with immense polymers of nearly unimaginable complexity, and a total atomic mass of well over 4,000,000 Daltons. By accretion, the ribosome has recorded its history, like the growth rings of a tree. The structure and integrity of ancient ribosomal components were buried and preserved during evolutionary construction of more modern elements. In our research we peel back the layers to map the evolution of lifes ancestral molecular machine. We create and test models of ribosomal origins and evolution, in silico, in vitro and in vivo, that make specific predictions of sequence, structure, folding, assembly, and catalysis. We experimentally test predictions by resurrecting ancestral ribosomes from various time points in ribosomal evolution.
May 15 2014 14:30	Fondazione Ernesto Illy - Via Parisi 10 Trieste	Prof. Dr. Hans Verhagen (Senior Scientific Advisor 'Nutrition and Food Safety' National Institute for Public Health and the Environment (RIVM) The Netherlands)	State of the art in benefit risk analysis: Food and Nutrition
ABSTRACT:
May 06 2014 10:30	SISSA - room 128	Prof. Antonio Celani (ICTP, Trieste)	Making sense of smell
ABSTRACT: The ability of organisms to sense their chemical environment is nearly as old as life itself. From viruses to metazoans, chemosensing is key to essential behaviors such as nutrient scavenging, mating, communication and sociality. I will discuss two prominent examples of chemical sensing, chemotaxis in bacteria and olfactory search in insects, with particular emphasis on how the behavioral response is linked to the perception of the chemical landscape.
Apr 29 2014 14:30	room 132	Prof. Craig Zirbel (Department of Mathematics and Statistics, Bowling Green State University)	Inferring RNA 3D motifs from sequence
ABSTRACT:
Apr 03 2014 14:00	room 128	Felix Ritort (Universitat de Barcelona)	Single molecules: from force spectroscopy to molecular evolution
ABSTRACT:
Mar 25 2014 10:00	Room 128-129	Dr. Daniele Di Marino (Università di Roma La Sapienza)	Studying the Fragile X Syndrome Using Combined Computational and Experimental Approaches: a Complex and Fascinating Story
ABSTRACT:
Mar 25 2014 11:00	Room 128-129 (Aula Cinema)	Dr. Luca Bortolussi (Università di Trieste)	Computational Modelling for Systems and Synthetic Biology
ABSTRACT:
Mar 24 2014 11:00	Room A-132	Dr. Pascal Auffinger (Université de Strasbourg, CNRS, IBMC)	Non-covalent interactions in RNA systems
ABSTRACT:
Mar 07 2014 10:00	Room A-132	Prof. Michal Otyepka (University of Palacky (Repubblica Ceca))	Human Cytochrome P450 on Membrane, Implications for Drug Metabolism
ABSTRACT:
Jan 23 2014 10:00	Room A-132	Luca Bortolussi (Department of Mathematics and Geoscience, University of Trieste)	Learning stochastic processes from qualitative data
ABSTRACT: Mathematical modeling is the key approach to understand the dynamics of complex systems. Stochastic models, in particular, can account for the intrinsic variability of the phenomenon we wish to model, in domains as diverse as biological and ecological systems, social networks, ICT. An important step in the modeling process is model calibration using experimental observations. Available techniques assume that we can measure the variables of the process at different time instants, ad exploit such time-series to find the model parameters that best fit observations. However, it is not always easy or feasible to obtain such data. Think about the spreading of gossips in a (small) social environment: It is unlikely that we will be able to measure at different time instants how many people are actively spreading the gossip, how many do not know it, and so on. What is more likely is that we can find out bounds on the fraction of the population that knows the information at steady state, or approximately when people would stop spreading the rumor. Similarly, some data produced in biological experiments can be quantified with a lot of imagination, and only qualitative assessments may be done reliably. The problem we will discuss is to what extent one can reconstruct a model parameterization only from qualitative data of such kinds, leveraging on recent advances in machine learning and computer science.
Dec 17 2013 15:00	Room A-132	M. Rosa (S3, CNR Institute of Nanoscience, Modena, Italy and Department of Physics, University of Modena and Reggio Emilia, Modena, Italy)	Multi-step computational approach to DNA surfaces
ABSTRACT: The interaction of DNA molecules with hard substrates is of fundamental importance both for the study of DNA itself and for the variety of possible technological applications. Interaction with inorganic surfaces strongly modifies the structure and geometry of DNA molecule, together with its electrical properties. Hence, an accurate understanding of DNA behavior at interfaces is a fundamental step foreseeing new nanotechnology applications. This work sets the fundamentals for the simulation of entire DNA oligomers on gold surfaces in dry and wet conditions. A multi-step approach was adopted for the study of the interaction between DNA and the Au(111) surface, relying on ab-initio, classical molecular dynamics and docking simulations. Thanks to first principles calculations a new GolDNA-AMBER force field was parametrized, including dispersion interactions and polarization effects. Thanks to this force field we simulated self-assembled guanine and adenine monolayers on Au(111) in vacuo, assessing the robustness of the force field through the comparison with first principles calculations and experiments, and the adsorption of all nucleobases on the same substrate in aqueous conditions.
Nov 08 2013 14:30	Room A-132	Prof. Antonio Di Carlo (Università degli Studi Roma Tre)	Divide et impera: how to scale up classical MD models
ABSTRACT: Imagine considering a material aggregate as either a molecular system or a continuous medium, and wishing to relate the two representations. Assume that the fields entering the continuum description, such as deformation and stress, are adequately sampled on an array of positions, whose typical spacing H is enormously larger than the average intermolecular distance d. Associate with each of these macroscopic sampling positions an Andersen-Parrinello-Rahman (APR) cell, whose reference size h is large enough with respect to d in order to allow for a decent sampling of the microscopic molecular states, and still much smaller than H: H >> h >> d. Now, let the molecules in each cell interact directly with each other (and with their h-neighboring images), while being indirectly affected by those in the H-neighboring cells via the collective degrees of freedom of the deforming APR cell, governed by the force balance and compatibility equations of the continuous medium (sampled at the H scale). In turn, the stress-deformation relation characterizing the response of the continuous medium arises as an emergent property of molecular dynamics on the h scale. Based on joint ongoing work with Manuela Minozzi and Matteo Paoluzzi, I will present and discuss a method for playing this sort of multiscale game.
Oct 04 2013 14:00	Room A-132	Dr. Enrico Rennella (Institut de Biologie Structurale, Université Grenoble)	Real-Time NMR Characterization of Structure and Dynamics in Transiently-Populated Protein Folding Intermediates
ABSTRACT:
Oct 04 2013 11:30	Room A-132	Hue Sun Chan (University of Toronto, Toronto, Ontario)	Cooperativity, Barriers, Nonnative Interactions, and the Diffusion Picture of Protein Folding
ABSTRACT: The Levinthal paradox of protein folding is commonly perceived as a statement about the impossibility of folding by a completely random conformational search. Often missed in such narratives is the fact that the question raised by Levinthal was in response to the experimental discovery of two-state, switch-like cooperative folding in the late 1960s, rather than to the problem of conformational search per se. The implication of this understanding on the notion of a funnel-like energy landscape will be discussed. Comparisons between theory and experiment on cooperative folding indicate a prominent role of desolvation barriers. Investigations into the role of desolvation in protein folding also resolves an apparent inconsistency between experimental observations of enthalpic folding barriers and the theoretical funnel picture of folding. Examples will be given to illustrate how important folding principles have been gleaned from studies using native-centric models, including a critical assessment of the diffusion perspective of folding and the concept of preequilibrium, and how nonnative interactions may be treated as a perturbation in essentially the same theoretical framework.
Aug 26 2013 14:00	Room A-132	Dr. Pilar Cossio (Max Planck Institute Frankfurt)	Linking simulation models and EM images through Bayesian probability measures
ABSTRACT:
Aug 01 2013 14:30	Room 132	G. DAdamo (Roma "La Sapienza" )	Consistent and transferable coarse-grained model for polymer solutions
ABSTRACT: I present a coarse-grained model for linear polymers with a tunable number of effective atoms (blobs) per chain interacting by intra- and inter-molecular potentials obtained at zero polymer density. I show how this model is able to accurately reproduce the universal properties of the underlying solution of athermal linear chains at various levels of coarse-graining and in a range of chain densities which can be widened by increasing the spatial resolution of the multiblob representation, i.e., the number of blobs per chain. The present model is unique in its ability to quantitatively predict thermodynamic and large scale structural properties of polymer solutions deep in the semidilute regime with a very limited computational effort, overcoming most of the problems related to the simulations of semidilute polymer solutions in good solvent conditions. I also present results concerning polymer systems in the good-to-theta solvent crossover regime. Also in this case it is possible to define a model which is simultaneously transferable with respect to the number of blobs, density and temperature
Jul 22 2013 16:00	Room A-132	Dr. Fabio Trovato (Scuola Normale Superiore di Pisa)	Complexity of bio-systems: (reduction by) coarse-grained modeling
ABSTRACT: One-bead per residue coarse-grained models of DNA/RNA and proteins (Green Fluorescent Proteins) are presented and shown to reproduce structural transitions like DNA supercoiling and RNA folding, GFP internal fluctuations, binding modes and diffusion in dilute and dense solutions. A careful parameterization and optimization of the CG force field, with a combination of statistics based (Boltzmann's inversion), physics and structure-based information by means of a Genetic Algorithm, allow to re-integrate directly experimental and theoretical data about the problem addressed. Pushing the time/length scales to the sub-ms/um domain requires additional simplifications, i. e. to treat an entire protein as a single interactive center. By adopting such a meso-scale resolution level, molecular dynamics simulations of the diffusion of 12 species of proteins present in the E. Coli cytoplasm, the nucleoid and the GFP are performed. Emphasis is put on the non-specific attractive intermolecular interactions, that together with volume excluded effects, account for the accurate matching with experimental data on diffusion and anomalous coefficients. The simplicity and transferability of the cytoplasm model is amenable to investigate the stability of biopolymers in conditions similar to the ones found in the cell.
Jul 11 2013 14:30	Room 005	Prof. Dr. Kurt Kremer (Max-Planck Institut für Polymerforschung, Mainz, Germany)	Topological Constraints Matter: Elastomers, Collapsed Polymer Globules, Chromosome Territories
ABSTRACT: The role of topological constraints on conformational as well as relaxational and dynamical properties of open linear and closed ring polymers as well as mixtures thereof is discussed. In the case of polymer melts the conformational statistics can be used to directly determine the entanglement molecular weight in excellent agreement to experiment. By manipulating the entanglements in long chain melts materials with new rheological properties can be achieved. For ring polymers the situation is completely different. While linked rings act like DeGennes Olympic gels, we find that non concatenated polymer rings segregate and form individual collapsed objects. I discuss some details of their conformations, which not only is related to one of the very basic problems in polymer science but also has far reaching consequences from the collapse of gels to chromosome territories.
Jul 10 2013 14:30	Room 005	Dr. Raffaello Potestio (Max Planck Institute for Polymer Research, Mainz, Germany)	Hamiltonian dual-resolution simulations of complex molecular liquids
ABSTRACT: Many phenomena occurring in soft matter, from the interaction of biomolecules to the self-assembly of biological as well as artificial nanocomposites, cover a broad range of length and time scales. Being hardly amenable with a fully-atomistic approach, these systems are usually addressed with computationally efficient coarse-grained models that, on the other hand, in many cases suffer from the lack of important chemical details. To circumvent this problem, adaptive dual-resolution simulation schemes have been devised to concurrently use different levels of resolution in different regions of the same system, yet allowing molecules to freely diffuse across the simulation domain. Here a method is presented, the Hamiltonian Adaptive Resolution Simulation (H-AdResS) scheme, which, in contrast to previous approaches, is built in terms of a Hamiltonian function; this makes it possible to formulate a solid statistical physics theory of dual-resolution systems and, as a consequence, it allows one to use the preferred statistical ensemble as well as the simulation algorithm.
Jun 17 2013 10:00	Room A-132	Louise Wright (Warwick, Chemistry Dept)	Facet Selectivity of AuBP-1: A Replica Exchange with Solute Tempering Metadynamics Study
ABSTRACT: Key to fully exploiting the unique physical properties of gold nanoparticles (AuNP) in future applications is the fine control over their size, shape and proximity on assembly. Biomimetic synthetic protocols, harnessing the selectivity of biomolecule binding observed in Nature, offer great potential in enabling us to reach this goal. In particular, peptide sequences able to distinguish between the different crystallographic planes of gold could be used to facilitate shape-selective nanoparticle synthesis. In order to rationally design such sequences, the mechanisms governing facet-selective peptide adsorption at the aqueous gold interface must be fully understood. Here, AuBP-1 [1] is used as a model system to investigate differential adsorption between Au(111) and Au(100)--the two surfaces most predominantly featured by AuNP. The reconstructed status of an Au(100) NP facet is currently unknown and so adsorption at both native and reconstructed interfaces is modelled. Advanced sampling techniques are employed to enhance conformational sampling of the bound peptide. Preliminary results indicate that binding to the Au(100)(1x1) surface is mediated by the interfacial water layer while AuBP-1 makes direct contact with gold at the other two interfaces. Trends in the free energy of adsorption suggest that adsorption to Au(111) is stronger than that to Au(100).
Jun 07 2013 10:00	Room A-132	Dr. Simon Poblete (Forschungszentrum Juelich GmbH)	Mesoscale Simulations of Multi-Domain Protein Dynamics
ABSTRACT:
Jun 06 2013 11:00	Room A-132	Areejit Samal (ICTP Trieste)	Phenotypic constraints drive the architecture of biological networks
ABSTRACT: Biological networks have architectural features that distinguish them from random networks such as degree distribution, high clustering, high robustness and over-representation of certain network motifs. In this talk, I ask whether such unusual features follow from phenotypic constraints on specific networks. The standard benchmark ensemble or null model used to detect unusual features in biological networks is the edge-randomization algorithm. I will present our new method based on Markov Chain Monte Carlo (MCMC) sampling to generate realistic benchmark ensembles for metabolic networks and gene regulatory networks with a given phenotype. I then redefine the notion of unusual features of biological networks using this realistic benchmark ensemble. I consider E. coli metabolic network and Boolean gene regulatory model for Arabidopsis flowering as two example systems from different levels of biological organization. By generating realistic benchmark ensembles for the two systems, I show that most unusual structural properties of these networks could arise due to functional constraints. I conclude that function is a main driver of biological network structure.
May 21 2013 14:30	Room A-132	Dr. Adolfo Poma (Università di Roma La Sapienza)	Assessing the quality of transmembrane protein models: A free energy study
ABSTRACT: Proteins mediate several processes in biological membranes, such as the exchange of molecules, energy, and information across the barrier that surrounds cells and organelles. In the pharmaceutical industry it is know that half of the drugs available in the market target to membrane proteins. Despite its importance, a unique technique able to predict the structure of transmembrane proteins is not yet accessible and as a consequence the database of one single protein might reach large sizes. Thus, a systematic method of assessment of the structural stability of the predicted models is needed. In this talk, we present a strategy to obtain free energy difference between models using restraint molecular dynamics (MD). By choosing the proper set of collective variables (CVs) we are able to explore the conformational space of models. Exploratory runs using temperature accelerated MD (TAMD[1]) using the same set of CVs points towards a metastability between predicted (NMR) models of the transmembrane part of the nicotine acetylcholine receptor (-nAChR-) beta2 subunit[2]. [1] L. Maragliano and E. Vanden-Eijnden, Chem. Phys. Lett., 426, 168-175 (2006). [2] Bondarenko V., Tillman T., Xu Y., Tang P. Biochimica et Biophysica Acta - Biomembranes, 1798 (8), 1608-1614, (2010).
May 20 2013 14:00	Room 005	Prof. Christoph Dellago (Faculty of Physics, University of Vienna)	Neural networks for energy calculation and structure recognition
ABSTRACT: Molecular dynamics simulations of nucleation phenomena in condensed matter systems require the accurate calculation of the forces acting on atoms. While ab initio approaches can be used for this purpose, their high computational cost often precludes their application to systems of adequate size. In this talk, I will report on a potential energy surface developed for copper sulfide based on the neural-network method of Behler and Parrinello. The neural-network potential has the accuracy of the density functional theory calculations, which have been used to generate reference data for the training of the network, at a fraction of the cost. Molecular dynamics simulations performed with this potential show the transition from the low-chalcocite to the high-chalcocite phase observed in experiments. Motivated by the success of neural network potentials we have also explored the application of this approach for the recognition of atomic arrangements. The ability to distinguish between different local structures is of great importance in the study of many condensed phase processes including freezing, structural phase transitions and the dynamics of defects. Several approaches to determine and classify the environments of individual atoms are currently available, among which local bond order parameters based on spherical harmonics and common neighbor analysis are the most popular ones. For complex substances with rich phase diagrams such as water, however, these methods fail to discriminate between different crystal structures. In this talk, I will show how neural networks can be trained to classify local structures in a simple and accurate way. We demonstrate the validity and practical applicability of the approach using Lennard-Jonesium and water/ice as illustrative examples. Joint work with Jörg Behler, Andreas Singraber, and Philipp Geiger
May 17 2013 11:00	Room 005	Prof. Hannes Jonsson (University of Iceland)	Adaptive Kinetic Monte Carlo method for Long Time Simulations and Global Optimization
ABSTRACT: By searching in an unbiased way for possible thermal transitions from a given state, without preconceived notion of final states or transition mechanism, the long time scale evolution of a system can be simulated using the adaptive kinetic Monte Carlo algorithm [1]. Systematic coarse graining of the free energy landscape is important to avoid being slowed down by fast processes. Within the harmonic approximation to transition state theory (HTST), the challenging task is to find all relevant saddle points on the energy rim surrounding an initial state energy minimum [2]. More generally, within full TST, a high dimensional dividing surface needs to be constructed and optimized using Keck's variational principle. In either case, the exact dynamics can, in principle, be obtained from short time trajectories started at the transition state [3]. This approach has been implemented to large extent in the distributed computing software EON (http://theochem.org/EON) [4] and applications to various atomic and spin systems will be described. A modification of the algorithm can be used for global optimization of objective functions of continuous variables [5]. Applications to metal cluster and geothermal reservoir modeling will be discussed. [1] 'Long time scale kinetic Monte Carlo simulations without lattice approximation and predefined event table', G. Henkelman and H. Jónsson, J. Chem. Phys., Vol. 115, p. 9657 (2001). [2] 'Efficient Sampling of Saddle Points with the Minimum-Mode Following Method', A. Pedersen, S.F. Hafstein and H. Jónsson, SIAM Journal of Scientific Computing 33, 633 (2011). [3] 'k-dynamics: An exact method for accelerating rare event classical molecular dynamics', C-Y. Lu, D.E. Makarov and G. Henkelman, J. Chem. Phys. 133, 201101 (2010). [4] 'Distributed Implementation of the Adaptive Kinetic Monte Carlo Method', A. Pedersen and H. Jónsson, Mathematics and Computers in Simulation 80, 1487 (2010). [5] 'Simulated Annealing with Coarse Graining and Distributed Computing', A. Pedersen, J-C. Berthet and H. Jónsson, Lecture Notes in Computer Science 7134, 34 (2012).
Apr 12 2013 11:00	Room 005	Prof. D. Amati; Prof. A. Laio; Dr. M. Maieron (SISSA)	A new method for brain fmri data analysis that allows signal identification on a single trial basis
ABSTRACT:
Mar 19 2013 16:30	Room A-132	Dott.ssa Elisa Frezza (Università di Padova)	Modelling of chirality propagation in self-assembling systems
ABSTRACT: Hierarchical self-assembly offers new strategies to build new complex materials [1, 2] and is characterised by molecular building blocks which form intermediate structures that self organise at macroscopic level. In nature there are a lot of remarkable examples of these processes (i.e. DNAs [3] or viruses) [4]; therefore, it is important to understand the mechanism to design and control molecular architectures for the purpose of building structures with desired properties and morphologies. It is needed to understand how the shape of building block influences the self-assembly [5, 6]. In this context, the chirality plays a crucial role: The chirality is extremely sensitive to subtleties on the molecular scale and can guide the self-assembly; furthermore, the chirality can act as amplifier of changes that occur at the molecular level [7, 8]. From the theoretical point of view, the difficulty of studying these phenomena derives from the need to develop multiscale methods and models, that can connect the different length scales. To take into account the relationship among the building blocks, their supramolecular organisation, and the properties of the aggregates, we need a detailed representation of intermolecular interactions; this description has to be integrated in a model of the collective behaviour of the systems. In this talk, I will present some topics related to chirality propagation in self-assembling systems and the theoretical and computational methods and model I used to understand their behaviour: (i) chiral liquid crystal phases of DNA [9], (ii) the phase diagram of hard helices [10] and (iii) self-assembly of conjugates porphyrin-peptide. [1] F. A. Aldaye, A. L. Palmer, and H. F. Sleiman Science 321, 1795 (2008) [2] M. A. C. Stuart, et al. Nature Materials 9, 101 (2010) [3] G. Zanchetta, F. Giavazzi, M. Nakata, R. Cerbino , N. A. Clark, and T. Bellini P. Natl. Acad. Sci. USA 107, 17497 (2010) [4] I. W. Hamley, Soft Matter 6, 1863 (2010) [5] P. Damasceno, M. Engel, and S. C. Glotzer, Science 337, 453 (2012) [6] J. de Graaf, R. van Roij, and M. Dijkstra, Phys. Rev. Lett. 107, 155501 (2011) [7] V. Schaller, and A. R. Bausch Nature 481, 268 (2012) [8] T. Gibaud, E. Barry, M. J. Zakhary, M. Henglin, A. Ward, Y. Yang, C. Berciu, R. Oldenbourg, M. F. Hagan, D. Nicastro, R. B. Meyer, and Z. Dogic Nature 481, 348 (2012) [9] E. Frezza, F. Tombolato, and A. Ferrarini Soft Matter 7, 9291 (2011) [10] E. Frezza, A. Ferrarini, H. B. Kolli, A. Giacometti, and G. Cinacchi J. Chem. Phys. submitted
Feb 04 2013 14:00	Room A-132	P. Ziherl (Faculty of Mathematics and Physics and Jozef Stefan institute , University of Ljubljana)	Structure and shape of animal tissues
ABSTRACT: The morphological diversity of animal cells is truly immense, and thinking of a unified theory to describe their mechanical properties is unlikely to bear any fruit. Yet there are many examples of tissues and cell aggregates exhibiting some kind of regularity which could be related to a more or less transparent physical mechanism - be it a packing constraint, a dynamical process, or an energy to be minimized. We review several equilibrium models of animal cell shapes based on bending, adhesion, and surface energies. In particular, we describe the structure of one-cell-thick epithelia which can be viewed as a polygonal tiling of a plane. Also discussed is a model of adhering polyhedral space-filling vesicles whose predictions are consistent with the structure of some types of mammal epidermis. We introduce a 2D theory of the formation of ventral furrow in the fruit fly, arguing that generic mechanics may play an important role in the embryonic development. Finally we use a similar model to describe the flat and the corrugated types of supported simple epithelia.
Feb 01 2013 11:00	Room A-132	Prof. Rosalind Allen (Università di Edinburgo)	History and chance in the development of microbial ecosystems
ABSTRACT:
Dec 03 2012 14:00	room 132	Dott. Andrea Spitaleri (Dulbecco Telethon Institute Center of Genomics, BioInformatics and BioStatistics, San Raffaele di Milano)	Molecular mechanism study of the encountering complex between the histone tails with PHD finger proteins: the effect of Epigenetic modifications
ABSTRACT:
Nov 16 2012 14:00	room 132	Davide Baù (Structural Genomics Team, CNAG - Center for Genomic Regulation (CRG), Barcelona (Spain))	Genome structure determination via 3C-based data integration by the Integrative Modeling Platform
ABSTRACT: The three-dimensional (3D) architecture of a genome determines the spatial localization of regulatory elements and the genes they regulate. Thus, elucidating the 3D structure of a genome may result in significant insights about how genes are regulated. The current state of the art in experimental methods, including light microscopy and cell/molecular biology, are now able to provide detailed information on the position of genes and their interacting partners. However, such methods by themselves are not able to determine the high-resolution 3D structure of genomes or genomic domains. We have developed a computational module of the Integrative Modeling Platform (IMP, http://www.integrativemodeling.org) that uses chromo- some conformation capture data to determine the 3D architecture of genomic domains and entire genomes at unprecedented resolutions. This approach, through the visualization of looping interactions between distal regulatory elements, allows characterizing global chromatin features and their relation to gene expression.
Oct 30 2012 11:00	room 132	Marco Cosentino-Lagomarsino (Universite' Pierre et Marie Curie, Paris)	Statistical physics and emergent laws of genome composition.
ABSTRACT: Quantitative approaches to evolutionary genomics, systems biology, and ecology unravel several universal regularities connecting genome-scale observables, phenotypes and physiological traits. While a complete physical theory of evolutionary biology is well beyond the reach of current science, some of these universals might qualify as "biological laws", similarly to how "law" is understood in modern physics. A current challenge for theoreticians is understanding how different universals be accounted for by mathematical models that capture their essential features. I will discuss large-scale data regarding the partitioning of all sequenced bacterial genomes into functional categories and evolutionary families, and show that their statistics follow well-defined "universal" laws. Finally I will discuss how some of these laws can be understood in terms of stochastic models that account for the basic "moves" available to a genome over evolution.
Oct 24 2012 15:00	room 132	Dr. Calin Floare (National Institute for R & D of Isotopic and Molecular Technologies - Cluj-Napoca-Romania)	Cyclodextrin inclusion compounds. Investigation of an unusual phase transition
ABSTRACT:
Oct 18 2012 14:00	room 132	Dr. Sandro Bottaro (SISSA)	Implicit solvent model parameterization via relative entropy minimization
ABSTRACT:
Oct 16 2012 16:00	room 132	Dr. Pasquale Pisani (Istituto Italiano di Tecnologia - IIT)	Application of Advanced Statistical and Machine Learning Methods To Computer Assisted Drug Design
ABSTRACT:
Jul 19 2012 14:30	room 004	Gianluca Lattanzi (Universita di Bari)	Non-equilibrium MD simulations of proteins and membrane interfaces
ABSTRACT:
Jul 18 2012 14:30	room 4	Prof. Mark Tuckerman (New York University)	Exploring free energy landscapes of peptides and crystalline polymorphs
ABSTRACT:
May 29 2012 11:00	room 132	Dott. Fabio Pietrucci (CECAM EPFL Lausanne)	Exploring complex transformation pathways in high-dimensional free energy landscapes
ABSTRACT: Many important physico-chemical processes "live" in a high-dimensional free energy landscape, which makes difficult their study by computer simulations. I will present some recently-developed methodologies which tackle this challenge: 1) Social PeRmutation INvarianT (SPRINT) topological coordinates, which in combination with ab-initio metadynamics allowed to discover the mechanism of the experimentally-observed transformation of a graphene flake into fullerene, and 2) conformational cluster models of protein dynamics built from all-atom bias-exchange simulations, applied to the early unfolding events of protein SH3 (compared with conventional replica exchange and with experiments) and to the full folding pathway of the WW domain.
May 28 2012 12:00	room 132	Prof. Reidun Twarock (York University, UK)	Viruses and Geometry: Where Symmetry Meets Function
ABSTRACT: In this talk we will discuss how symmetry properties of viruses impact on their function. In particular, we will show that the structure of the packaged genome implies an assembly scenario for the protein container, via RNA-capsid protein interactions, that is evolutionarily conserved in a viral family. We will moreover show how symmetry properties of the capsid impact on the structural transitions of viruses, in particular plant viruses, that are important for infection.
May 21 2012 15:00	room 132	Dott. Tatjana Skrbic (Università di Trento)	Protein folding: from coarse-grained to atomistic simulations
ABSTRACT:
May 18 2012 11:00	room 132	Dott. Omar Vallson (University of Twente, Netherland)	Native Description of Visual Absorption: A Challenge for First Principles Calculations
ABSTRACT:
May 08 2012 11:00	room 132	Dott. Maurizio Rossi (Università di Milano)	Study of solid 4He in 2D: Defected and Confined Crystal
ABSTRACT:
Apr 12 2012 11:00	room 132	Dott. Ivan Gladich (Academy of Sciences of Czech Republic)	Arrhenius Analysis of Anisotropic Surface Self-Diffusion on the Prismatic Facet of Ice
ABSTRACT:
Mar 19 2012 11:00	room 132	Dott. Sandro Bottaro (Technical University of Denmark (DTU) )	Conducting Monte Carlo simulations of dense molecular systems
ABSTRACT: Although Markov chain Monte Carlo simulation is a potentially powerful approach for exploring conformational space of biomolecules, it has been unable to compete with molecular dynamics in the analysis of high density structural states, such as the native state of globular proteins. Here, we introduce a novel Monte Carlo move, CRISP, that greatly enhances the sampling effciency in all-atom MC simulations of dense systems. The algorithm is based on an exact analytical solution to the classic chain-closure problem, making it possible to express the interdependencies among degrees of freedom in the molecule as correlations in a multivariate Gaussian distribution. We demonstrate that our method reproduces structural variation in proteins with greater efficiency than current state-of-the-art Monte Carlo methods, and has real-time simulation performance on par with molecular dynamics simulations. The presented results suggest our method as a valuable tool in the study of molecules in atomic detail, offering a potential alternative to molecular dynamics for probing long time-scale conformational transitions.
Mar 13 2012 10:00	room 132	Dr. Maria Darvas (Laboratory of Interfaces and Nanosize System Institute of Chemistry ELTE University, Budapest, Hungary)	Anesthetic Molecules Embedded in a Lipid Bilayer
ABSTRACT:
Mar 06 2012 14:00	room 132	Dr. Rok Borstnar (National Insttitute of Chemistry, Lubiana)	Can Active Center pKa values of Monoamine Oxidase B give insight into the Mechanism of the Enzyme?
ABSTRACT:
Mar 01 2012 14:30	room 132	Dr. Davide Branduardi (Max Planck Institute for Biophysics . Frankfurt am Main)	Metadynamics with flexible hills shape
ABSTRACT: Metadynamics is a well known method for free energy calculations which has been successfully applied to many problems related with overcoming free energy barriers larger than KbT. Since this method relies on the addition of repulsive multidimensional Gaussian potentials, is it mandatory to choose their widths and orientation in each dimension. These widths are a measure of the resolution of the expected free energy while the Gaussian orientation is a measure of the linear dependence among different variables. Both parameters are generally kept constant during the simulations (with some exceptions). Here I will show that it is possible to introduce two simple algorithms for calculating hills widths and orientation on the fly by choosing only one parameter, despite the dimensionality of the calculation. By taking advantage of the welltempered metadynamics formulation, the correct estimate of the free energy is easily retrieved. I will show numerical example by using alanine dipeptide in vacuum.
Feb 23 2012 14:30	room 132	Fabrizio Marinelli (MPI Mainz)	Following easy slope paths on a free energy surface
ABSTRACT: A new method for the automatic exploration and computation of multidimensional free energy landscapes is presented. It relies on concepts derived both from metadynamics and minimum-mode following methods. Like metadynamics the approach uses a certain number of collective variables that are thought to be relevant for the process under investigation and then the exploration of this space is boosted by introducing a history dependent potential that enhance on time the exploration of not visited configurations. Inspired by minimum-mode following methods, the functional form of the bias potential is chosen in order to allow the system escaping a local free energy minimum following the direction of slow motions. This allows using a larger number of collective variables compared to methods that, like metadynamics, are not based on a specific direction of the biasing force. The new technique is first tested on Ala-Pro and Ace-Ala3-Nme peptides, then it is applied to the ab-initio folding of Trp-cage and Advillin c-terminal headpiece in explicit solvent. For both proteins, starting the simulation from completely unfolded conformations, the method allows to visit a wide range of conformations, including nearly native ones, within a few hundreds of ns. In the case of the Trp-cage, multiple independent simulation runs were also performed to achieve a detailed classification of different possible folding pathways. Based on these results, the new methodology successfully explores the relevant configurations of a multidimensional free energy landscape allowing to predict the main pathways pertaining to complex transitions like the folding of small proteins.
Feb 20 2012 14:30	room 132	Vanessa Leone (MPI Mainz)	Ion specificity and transport mechanism of the membrane motor of ATP synthase
ABSTRACT: ATPases are membrane-associated proteins that are able to couple ATP synthesis/hydrolysis to proton or sodium flow down or against their electrochemical gradients. The mechanism of these enzymes involves two opposing rotary motors, known as F1 and Fo in the F-type subfamily, which are physically connected by central and lateral stalks. F1 is where ATP is synthesized or hydrolyzed, by a mechanism that is reasonably understood. By contrast, much less is known about the membrane Fo complex, both in regard to its structure and the ion-transport rotary mechanism. In a close interplay between traditional biochemical approaches and computational techniques we are studying features of the Fo complex (or its equivalent) that have important mechanistic and functional implications. Here I will introduce two of these studies: one is the structural basis for the ion specificity of membrane rotors; second is the structure of the interface where ion binding and release take place, formed by the so-called c-ring and the subunit-a. In the first study, we built an assembly of a particular archaeal rotor by homology modelling and, subsequently, we compute the free energy of selectivity (Na+vs. H+) by atomistic simulations. The Na+ vs H+ selectivity obtained explains the unique behaviour of this ATPase at different pH and Na+ concentrations. In the second study we derive, upon existing biochemical data, plausible structural models through Rosetta ab-initio folding and docking algorithms. Based on these models, we speculate on the characteristics of the proton pathway across the membrane.
Feb 13 2012 12:00	room 132	Dott. Alessandro Mossa (Institut for Fysik og Astronomi, Aarhus Universitet, Denmark)	A novel fluctuation theorem to study misfolded and metastable states
ABSTRACT: The manipulation of individual macromolecules made possible by experimental techniques such as optical tweezers or atomic force microscopy gives a unique insight into the nonequilibrum thermodynamics of small systems. Besides a general introduction about the theoretical and experimental framework, this talk is focused on a powerful generalization of Crooks fluctuation theorem that allows the exploration of misfolded and metastable states. Even though all the examples I will show are applications to biophysics, the techniques are general and can be used for a vast range of problems in the soft matter field.
Jan 26 2012 11:00	room 132	Dott. Massimiliano Anselmi (Università La Sapienza, Roma)	Theoretical and computational study of globins in different environments
ABSTRACT: Globins small size, structural stability and rather complex functional behavior make such a type of proteins model systems to perform studies, at the atomistic level, on the relation between structure, dynamics and function. Globins, such as Myoglobin and more recently Neuroglobin, were extensively investigated in order to describe ligand diffusion in protein matrix, to identify escape routes and to characterize couplings between structural changes and properties relevant to biological activity. A large amount of data reported in literature was obtained in different condensed phases and in a wide range of conditions. Using theoretical and computational techniques, it should be possible to reproduce globins properties in different environments making more effective the comparison between several experimental data. Such comparisons are not often done because investigators are often more interested in studying the properties of a protein in aqueous solution loosely resembling physiological conditions or most experiments. Also it is usually assumed that the effect of crystal packing on the protein structure is small and therefore crystal structures look like the structure of a protein in solution. However crystal structures represent a confined medium that may introduce artifacts, due to the interactions of a protein complex with neighboring molecules. Molecular Dynamics simulations performed in different environments permitted to discover differences or discrepancies between experimental or computational data obtained either in solid or in fluid state.
Jan 24 2012 11:00	room 132	Lizhe Zhu (Van't Hoff Institute of Molecular Sciences, University of Amsterdam)	Modeling mechanisms of transmembrane signaling
ABSTRACT: Signal transduction happens at the cell surface as a signal ligand is caught by the transmembrane (TM) receptor. Such ligand binding introduces a change in the receptor protein and ultimately leads to cell responses. This change of the receptor could not only be a conformational change but also a change in the scale of its fluctuations. We discuss these two models of mechanisms via two receptor systems. HAMP domains are part of bacteria chemotaxis receptors crucial in relaying the signals. We find, via all-atom molecular dynamics and well-tempered metadynamics, that the prototype HAMP domain, when isolated, exhibits a collective motion yet with only one stable conformation. Using replica exchange Monte Carlo sampling of a simple model, we also show that integral membrane receptors, such as G-Protein-Coupled-Receptors, can work purely via fluctuations, under the condition that the receptor consists of at least four TM helices.
Dec 19 2011 16:00	Room 005	Dr. Felix Ritort (Small Biosystems Lab, Universitat de Barcelona, Spain)	Unraveling nucleic acids interactions by molecular unzipping
ABSTRACT: Recent developments in micro and nano-technologies allow for the controlled manipulation of individual molecules by exerting and detecting forces in the piconewton range. Molecular unzipping is a force-induced reaction that makes possible to disrupt the bonds that hold molecular structures in nucleic acids and proteins. In this way a double stranded DNA molecule can be converted into two individual single strands by pulling apart the two strands. The capability of single molecule techniques of detecting weak forces together with the ability of measuring extensions with nanometer resolution allow scientists to monitor molecular reactions in real time (e.g. molecular folding). In this talk I will review the most recent applications of molecular unzipping that make possible to derive base-pair free energies in DNA with unprecedented accuracy within tenths of a kcal/mol. I will then show the potentialities of molecular unzipping as a tool to detect structural transitions in DNA and to unravel DNA-peptide interactions.
Dec 14 2011 16:00	Room 132	Prof. Michele Vendruscolo (University of Cambridge)	Determination of the structures and dynamics of proteins using NMR chemical shifts
ABSTRACT:
Nov 30 2011 16:30	Room A-132	Prof. Edoardo Milotti (Dipartimento di Fisica Università di Trieste)	Numerical simulations of the growth of tumor spheroids
ABSTRACT: At present it is still quite difficult to match the vast knowledge on the behavior of individual tumor cells with macroscopic measurements on clinical tumors. On the modeling side, we already know how to deal with many molecular pathways and cellular events, using systems of differential equations and other modeling tools, and ideally, we should be able to extend such a mathematical description up to the level of large tumor masses. An extended model should thus help us forecast the behavior of large tumors from our basic knowledge of microscopic processes. Unfortunately, the complexity of these processes makes it very difficult -- probably impossible -- to develop comprehensive analytical models. We try to bridge the gap with a simulation program which is based on basic biochemical and biophysical processes -- thereby building an effective computational model -- and in this seminar I describe its basic structure.
Nov 23 2011 16:00	Room 132	Dr. Alex Rodriguez (Chemical engineering Department in the Technical University of Catalonia (UPC))	Using replica exchange molecular dynamics for exploring the free energy landscape of polypeptides: a few examples
ABSTRACT:
Nov 17 2011 16:00	Room 132	Dr. Davide Branduardi (Max Planck Institute for Biophysics, Frankfurt am Main (Germany))	Calculating free energy lanscapes of chemical reactions with path collective variables
ABSTRACT:
Oct 18 2011 10:00	Room 005	Prof. Gian Gaetano Tartaglia (CRG Barcelona)	Towards quantitative prediction in cell biology
ABSTRACT: We have recently developed a powerful algorithm to predict protein-RNA associations that we validate using a series of assays including immunoprecipitation of protein-RNA complexes. At the same time, we work on the prediction of interactomes of amyloid fibrils in the cellular context and develop models for calculating aggregation rate and the interaction potential with molecular chaperones. A series of bioinformatics tools to predict toxicity under a variety of experimental conditions will be presented.
Sep 26 2011 12:00	Room A-132	Dr. Kristen Marino (University of Amsterdam)	Confinement induced stable states in Trp-cage folding via all-atom simulations
ABSTRACT:
Jun 21 2011 11:00	Room 005	Prof. Massimiliano Di Ventra (Department of Physics University of California, San Diego )	Fast DNA sequencing: a physicists perspective
ABSTRACT: Fast and low-cost DNA sequencing methods would revolutionize medicine: a person could have his/her full genome sequenced so that drugs could be tailored to his/her specific illnesses; doctors could know in advance patients likelihood to develop a given ailment; cures to major diseases could be found faster. However, this goal of personalized medicine is hampered today by the high cost and slow speed of DNA sequencing methods. In this talk, I will first give an overview of recent proposals to achieve fast DNA sequencing using several techniques, ranging from optical to capacitive. I will finally discuss the sequencing protocol we suggest which would require the measurement of transverse currents during the translocation of single-stranded DNA into nanopores and support our conclusions with a combination of molecular dynamics simulations coupled to quantum mechanical calculations of electrical current in experimentally realizable systems. I will also discuss recent experiments that support these theoretical predictions. In addition to their possible impact in medicine and biology, the above methods offer ideal test beds to study open scientific issues in the relatively unexplored area at the interface between solids, liquids, and biomolecules at the nanometer length scale [1]. [1] M. Zwolak, M. Di Ventra, Physical Approaches to DNA Sequencing and Detection, Rev. Mod. Phys. 80, 141 (2008).
May 30 2011 11:30	Room A-132	Diego di Bernardo (TIGEM - University of Naples Federico II)	Reverse-engineering and modeling gene networks: from gene function to drug discovery
ABSTRACT: ne of the main aim of Systems Biology is inferring, or reverse-engineering, gene networks. This process can be defined as that of identifying gene interactions from experimental data through computational analysis. We will show how we used these techniques to identify the function of disease genes and to identify "master regulators" of gene signatures. In addition, we will also present published results on a new method to elucidate the mode of action of drugs. Conversely, Synthetic Biology allows de novo construction of a regulatory network to seed new functions in the cell. Synthetic biology aims to use such models to design unique biological circuits (synthetic networks) in the cell able to perform specific tasks (e.g., a "transcriptional clock" for periodic expression of a gene of interest) or to change a biological process in a desired way (e.g., modify metabolism to produce a specific compound of interest). In this framework, we will present our recent results on the construction of a mammalian transcriptional clock.
May 30 2011 14:00	Room A-132	Dr. Thomas Sexton (IGH CNRS - Montpellier)	Global mapping of genomic spatial interactions in the fruit fly
ABSTRACT: The nucleus is a highly heterogeneous organelle, with a non-random spatial arrangement of the genome and the factors that process it, such as the DNA replication and transcription machinery. At one level, it has been shown that regulatory DNA elements can exert effects on target genes over distances of hundreds of kilobases, and that these effects are often correlated with physical contact of the target gene. At a higher level, genes from the same and different chromosomes have been found to spatially cluster at nuclear foci that are enriched in their regulatory factors, indicating that groups of genes may be co-ordinately regulated via their 3-D organisation.
Mar 30 2011 14:30	Room A-132	Dr. Vittorio Limongelli (ETH Zurich)	Free Energy Landscapes in the Docking World
ABSTRACT: Continuos efforts are made by experimental and theoretical scientists to unravel the interactions between a ligand and its target protein thus understanding the way a protein functions and providing useful information for drug discovery. In an ideal situation, this information comes from high resolution crystal structures and ligand-docking protocols represent the elective tool to obtain correct information about the binding. However this standard procedure fails when either protein conformational changes or solvent effects play a role.
Mar 25 2011 14:30	Room A-132	Dr. Dariusz Ekonomiuk (University of Warsaw)	Computational structural biology against world-wide spread viral disease
ABSTRACT:
Feb 24 2011 14:30	Room 004	Dr. Stefano Piana (DE Shaw Research)	Towards accurate and quantitative atomistic simulations of biomolecules
ABSTRACT:
Feb 17 2011 10:00	Room A-132	Dr. Svea Sauer (Bremen Center for Computational Materials Science)	Solubility of ZnO Nanoparticles
ABSTRACT: Nanomaterials used in commercial products are in the focus of interest regarding their toxic properties at the nano-bio interface. The toxicity of ZnO nanoparticles towards mammalian cells was found to be directly related to particle dissolution. Furthermore a dependence of the solubility of zinc oxide nanoparticles on the surrounding solute could be observed. Therefore this project puts a focus on the dissolution of single Zn ions from zinc oxide surfaces and how this is influenced by the presence of different components of a cell culture media. As a starting point for all following research regarding the solubility, we look at the water adsorption on the surfaces of ZnO, as water is always present in a biological environment. Furthermore priliminary considerations about how to use Metadynamics in order to look at different dissolution scenarios are presented.
Feb 14 2011 14:00	Room 005	Giorgio Colombo (CNR - Milano)	Structure-function-dynamics relationships in proteins: implications for drug discovery
ABSTRACT: In this study, we will present recent results on the development of computational biology strategies for the discovery of new inhibitors of protein-protein interactions with drug-like properties, and for the study of the functional dynamics and allosteric signal propagation mechanisms in proteins.
Jan 26 2011 16:00	Room A-132	Dr. Guillaume Witz (EPFL, Lausanne, Switzerland )	Effects of topology on DNA conformations: AFM experiments and numerical simulations
ABSTRACT:

SBP Group Seminar History

REFERENCES