Upcoming seminars
Friday January 17, 2020, 11:00, Room C0.110 (UvA - Science Park)
Speaker: Prof. Nicola Marzari
EPFL (Lausanne, Switzerland)
Host: Evert Jan Meijer
Title:
The great acceleration in materials discovery
Abstract:
Materials are at the core of our technological advances, and are needed to address many of our societal challenges: from energy to information, from food to medicine. I'll highlight the great strides made in last few years in the design and discovery of novel materials, where computational simulations can now precede, streamline, or accelerate experiments. This acceleration is driven by the central paradigm of computational science (doubling performance every 14 months), combined with powerful and predictive quantum simulation techniques, and by the convergence of data mining and machine learning towards materials simulations. I'll also underscore the IT requirements needed to perform calculations in a reproducible, shareable, high-throughput mode, highlighting applications to novel two-dimensional materials and solid-state Li-ion conductors.
Previous seminars
Friday September 20, 2019, 11:00, Room D1.112 (UvA - Science Park)
Speaker: Prof. Rong-Zhen Liao
Huazhong University of Science and Technology (Wuhan, China)
Host: Evert Jan Meijer
Title:
Challenges and Progress in Modeling Water Oxidation Reactions
Abstract:
Water splitting driven by light is a promising technology for supplying clean and sustainable fuel by production of hydrogen. The oxidation of water by releasing four protons and four electrons is therm odynamically unfavorable with a relatively large energy demand (E0=1.23 V vs SHE at pH=0), and is therefore quite challenging to accomplish. During the last few decades, considerable progress has been achieved in the development of homogeneous water oxidation catalysts (WOCs) using transition metals. Quantum chemical calculations have been used to elucidate the mechanism of water oxidation promoted by a number of catalysts incorporating Ru, Ir, Fe, Mn, and Cu. In this talk, some of the challenges in the modeling are discussed.
Friday June 28, 2019, 11:00, Room D1.115 (UvA - Science Park)
Speaker: Dr. Roberto Covino
Max Planck Institute for Biophyics, Germany
Host: Peter Bolhuis
Title:
Artificial Intelligence-Assisted Discovery of Molecular Mechanisms from Computer Simulations
Abstract:
Computer simulations are a powerful instrument to discover reaction mechanisms of molecular events. Increasing computational power and smart software enable us to simulate longer time scales, resulting in ever larger data in the form of trajectories. Our ability to quantitatively interpret these data, however, lags behind, because it still almost completely relies on careful visual inspection by human experts. We need to overcome this interpretation bottleneck to manage and take advantage of the data-avalanche expected in the Exascale era. Obtaining a mechanistic understanding is not only the ultimate goal of many simulations but also enables sophisticated enhanced sampling schemes, which give access to time scales that cannot be sampled by straightforward simulations. I will introduce a new computational framework that combines deep learning, transition path sampling simulations, and statistical inference to accelerate the simulation of interesting molecular events and simultaneously learn their mechanism. I will report on solutions to make the trained neural networks interpretable by distilling simple reaction coordinates from the simulated data. This new computational framework represents a step forward towards a fully autonomous data driven production and interpretation of computer simulations of molecular events.
Friday February 23, 2018, 11:00, Room C1.112 (Science Park)
Speaker: Dr. Martin Engler
University of Düsseldorf, Germany
Host: Ivan Kryven
Title:
Graph-theoretical approaches to molecular parameterization for MD simulations
Abstract:
MD simulations at an atomic or near atomic level are playing an ever increasing role. The accuracy of MD simulations is determined by its atomic interaction parameters (such as point charges, bond lengths, bond angles and torsion angles). Over the last two decades, several automated parameterization tools have been developed which rely on Quantum Mechanical (QM) calculations and/or on sets of empirical rules. While those approaches are feasible for simple organic molecules (< 40 atoms), computing parameters for larger molecules which cannot be represented as combinations of simple sub-units remains a major challenge. In fact, the runtime and accuracy of these calculations scale quite poorly with the size of the molecule. A potential solution to this problem is to use a "fragment-based" approach. That is to generate a topology of a target molecule from a series of overlapping fragments obtained by matching the target to an existing database of parameterized smaller molecules. This approach requires efficient algorithms to enumerate common molecular substructures. In this talk, we present an efficient algorithm for enumerating maximal common fragments (MCF-E, a generalization of the problem of enumerating common connected induced subgraphs) and its applications including a new web-based tool for computer-assisted molecular parameterization.
Wednesday January 17, 2018, 09:00, Room C0.110 (Science Park)
Speaker: Prof. Gregory Voth
Department of Chemistry, The University of Chicago, USA
Host: Peter Bolhuis
Title:
Systematic Coarse-graining: Fundamentals and Applications
Abstract:
Coarse-grained (CG) models provide a computationally efficient means to study biomolecular and other soft matter processes involving large numbers of atoms correlated over distance scales of many covalent bond lengths and long time scales. Variational methods based on information from simulations of finer-grained (e.g., all-atom) models – for example the multiscale coarse-graining (MS-CG) and relative entropy minimization methods – provide attractive tools for the systematic “bottom up” development of CG models. These methods can also be extended to the “ultra coarse-grained” (UCG) regime, e.g., at a resolution level coarser or much coarser than one amino acid residue per effective CG particle in proteins. This extension leads to the possible existence of multiple metastable states “within” the CG sites for a given UCG model configuration, which can also be treated by systematic variational methods. Additionally, certain aspects of this theory connect back to single-state force matching and open up new avenues for method development in that arena. These results taken as a whole provide a formal statistical mechanical basis for CG methods related to force matching and relative entropy minimization and suggest practical algorithms for constructing optimal CG models from fine-grained simulation data, or for utilizing CG models to bias new fine-grained simulations and to enhance their sampling. Representative applications to challenging liquid and biomolecular systems will be described.
Friday November 7, 2014, 11:00, Room B0.207 (Science Park)
Speaker: Prof. dr. Frank Noe
Computational Molecular Biology group in the Freie Universität Berlin, Germany
Host: Peter Bolhuis
Title:
Efficiently extracting thermodynamics and kinetics from molecular simulation data at multiple thermodynamic states
Abstract:
I will present novel methods based on Markov modeling for extracting statistical information (thermodynamics and kinetics) from molecular simulation data that has been generated at multiple thermodynamic states. Such data may be obtained from enhanced sampling protocols, such as umbrella sampling or replica-exchange dynamics, and by mixing one of these protocols with direct molecular dynamics data. Here I will propose ways to optimally extract information from such data, including the reconstruction of the kinetics of rare events that are not directly sampled in the data. An application of our approach is the estimation of rare unbinding kinetics of protein-ligand complexes when only the more frequent binding process can be sampled in direct MD simulations.
Friday March 13, 2015, 11:00, Room G3.02 (Science Park)
Speaker: Prof. dr. Philippe Sautet
Ecole Normale Supérieure de Lyon and the Université de Lyon, Lyon, France
Host: Evert Jan Meijer
Title:
Modelling materials and reactions for energy conversion and storage
Abstract:
Chemical energy conversion and storage requires the design of novel materials, as catalysts or storage media. First-principles atomistic simulation is today a key tool to understand better the properties of these materials, to design in-silico potential candidates, and to understand their specific catalytic reactivity towards molecules. In this lecture I will discuss first the simulation of key properties of photocatalytic materials, and second the modelling of reaction pathways for the electrocatalytic transformation of CO2.
Photocatalytic systems, such as those used for hydrogen production from water generally imply at least one semiconductor in their architecture which role is to absorb the light or to transport the charge carriers. Despite the large variety of working principles encountered in these systems, they share some fundamental steps such as light absorption, exciton dissociation and charge carrier diffusion. The key properties of the semiconductor are the bandgap, the dielectric constant, the charge carrier effective masses and the exciton binding energy. Can these properties be accurately calculated or predicted from first principle calculations? [1] This is the first question I will address, with specific applications to doped TiO2 materials and tantalum (oxy)nitride compounds [2-4]. The second topic will bring us to electrocatalysis for transformation of CO2 in formic acid and vice versa. HCOOH formation is the formally simplest case of CO2 recycling, requiring only two electrons and two protons while leaving the CO2 skeleton intact. Formic acid can be used as a safe and convenient equivalent of H2 in transfer hydrogenations. Moreover, the rather non-hazardous liquid has a high volumetric hydrogen density and could be used in direct fuel cells. Here we investigate the kinetics of the formation and decomposition of formic acid under electrochemical conditions. Despite recent progress in modelling electrocatalytic reactions, there is still no consensus on how to assess the effect of the electrode potential at an atomic scale or to adequately model the influence of the solvent. With an explicit account of the electrochemical potential and an implicit solvent model, we explore the electrocatalytic reaction pathways in detail on Ni(111) and demonstrate that the potential dependence of reaction barriers should not be neglected, even for "innocent" looking reorientations of adsorbed formate. We also show that desorption/adsorption of CO2 is activated and that this barrier can have an important impact on the overall energy profile, and hence on the kinetics.
References
1) Le Bahers T, Rérat M, Sautet P J. Phys. Chem. C 2014; 118: 5997-6008.
2) Harb M, Sautet P, Raybaud P J. Phys. Chem. C 2011; 115: 19394.
3) Nurlaela E, Ould-Chikh S, Harb M, del Gobbo S, Aouine M, Puzenat E, Sautet P, Domen K, Basset J- M, Takanabe K Chem. Mater. 2014; 26: 4812.
4) Harb M, Sautet P, Nurlaela E, Raybaud P, Cavallo L, Domen K, Basset J-M, Takanabe K Phys. Chem. Chem. Phys. 2014; 16: 20548.
Friday September 26, 2014, 11:00, Room C4-174 (Science Park)
Speaker: Prof. dr. Gerhard Hummer
Department of Theoretical Biophysics, Max Planck Institute of Biophysics, Frankfurt, Germany
Host: Peter Bolhuis
Title:
Molecular simulation of protein dynamics and function
Abstract:
The combination of modern molecular simulations and quantum chemical computations makes it possible to study the molecular function of proteins in unprecedented detail. We have been using this combination to elucidate the molecular mechanisms and chemical reaction principles underlying the enzymatic catalysis of complex multistep reactions, with the processing of nucleic acids using two-metal ion catalysis as a paradigmatic example. A combination of quantum chemical calculations and molecular simulations also helped clarify the early time evolution of the molecular structure of the light-activated signaling protein PYP, in combination with picosecond time-resolved X-ray crystallography. On larger scales of space and time, simulations help us explore the motions and function of the molecular machines involved in biological energy transduction, including the F1 rotary motor ATP synthase and the proton pump Complex I. Remarkably, common physical principles emerge in the function of these proteins, despite large variations in their structure and function, including water and hydration effects, extended protein motions, and long-range electrostatic couplings.
Friday September 12, 2014, 15:00, Room C4-174 (Science Park)
Speaker: Prof. J. Ilja Siepmann
Department of Chemistry, University of Minnesota, USA
Host: Peter Bolhuis
Title:
Predictive Modeling of Separation Processes: Solvent Extraction, Adsorption, and Chromatography
Abstract:
Molecular simulations using efficient Monte Carlo algorithms and accurate force fields are used to explore complex separation processes: (i) the liquid-liquid extraction of ethanol from aqueous solution, (ii) the adsorption of binary and ternary mixtures of alcohols and water onto silicalite-1 and other zeolites, and (iii) the retention of alkanes, alcohols, and arenes in reversed-phase liquid chromatography. The simulations yield molecular-level information on the factors that govern selectivity and capacity and aid in the design of more efficient separation processes.
Tuesday April 15, 2014, 11:00, in S-607 (VU, W&N building)
Speaker: Dr. David Branduardi
Crucell, Leiden, the Netherlands
Host: Bernd Ensing
Title:
String method in cartesian space for freely-tumbling systems: ATP-Mg2+ conformational transitions
Abstract:
Very often the prediction of the long-time dynamics of a system requires the calcu- lation of the free-energy landscape of the process of interest. This can be generally obtained with straightforward sampling or with enhanced-sampling methods like umbrella sampling [1]. Nevertheless the dynamics obtained from these calculations may be strongly affected from artefacts that are corrected with the introduction of transmission coefficients, which are rather cumbersome to calculate. So there is the need of finding the best possible free energy profile that avoids this need.
String method [2] is a strategy that aims to find the best one-dimensional free-energy profile derived from a highly-dimensional descriptor space. In this respect Cartesian space if the most intuitive and complete descriptor space. However its use in prac- tice is affected by its non roto-translational invariance.
Here we introduce a novel flavour of roto-translational invariant string method in Cartesian space [3] which solves this issue and we show its application in the case of alanine dipeptide in vacuum and adenosine-5'-triphosphate (ATP) complexed with Mg2+ ion in water.
We investigate the main conformational barriers, analyse the committor probability and perform free energy decomposition. Interestingly we observe how microsolvation plays a key role that cannot be captured with traditional coordination parameters.
[1] Roux, B. "The calculation of the potential of mean force using computer simu- lations" (1995), Comp. Phys. Comm., 91, 275-282.
[2] Maragliano, L.; Fischer, A.; Vanden-Eijnden, E.; Ciccotti, G. "String method in collective variables: minimum free energy paths and isocommittor surfaces" (2006), J. Chem. Phys., 125(2), 024106.
[3] Branduardi, D.; Faraldo-G ́omez, J.D. "String method for calculation of minimum free-energy paths in cartesian space in freely tumbling systems" (2013), J. Chem. Theor. Comp., 9(9), 4140-4154.
Monday January 27, 2014, 11:00, Room D1.112 (Science Park)
Speaker: Dr. Anton Feenstra
IBIVU Bioinformatics, Vrije Universiteit, Amsterdam
Host: Jocelyne Vreede
Title:
Fast and accurate calculation of protein-protein interaction free energy
Abstract:
Motivation:
To assess if two proteins will interact under physiological
conditions, information on the interaction free energy is needed.
Statistical learning techniques and docking methods for predicting
protein-protein interactions cannot quantitatively estimate binding free
energies. Full atomistic molecular simulation methods do have this
potential, but are completely unfeasible for large-scale applications in
terms of computational cost required. Here we investigate whether
applying coarse-grained (CG) molecular dynamics simulations is a
viable alternative for complexes of known structure.
Results:
We calculate the free energy barrier with respect to the
bound state based on molecular dynamics simulations using both a
full atomistic and a CG force field for the TCR-pMHC complex and the
MP1-p14 scaffolding complex. We find that the free energy barriers
from the CG simulations are of similar accuracy as those from the
full atomistic ones, while achieving a speedup of over 500-fold. We
also observe that extensive sampling is extremely important to obtain
accurate free energy barriers, which is only within reach for the CG
models. Lastly, we show that the CG model preserves biological
relevance of the interactions: i) we observe a strong correlation
between evolutionary likelihood of mutations and the impact on the
free energy barrier with respect to the bound state; and ii) we
confirm the dominant role of the interface core in these interactions.
Our results therefore suggest that CG molecular simulations can
realistically be used for the accurate prediction of protein-protein
interaction strength.
Friday January 17, 2014, 11:00, Room C0.05 (Science Park)
Speaker: Prof. Ralf Everaers
Ecole Normale Supérieure de Lyon, France
Host: Evert Jan Meijer
Title:
Universal aspects of chromosome folding
Abstract:
Human chromosomes are fiber-like complexes of histone proteins condensing DNA double helices of ~ 10^8 base-pairs. The dynamics of the mm-long chromatin fibers in the cell nucleus is subject to strong topological constraints. In particular, their incomplete equilibration during interphase results in territorial, entanglement-free crumpled globule-like chromosome conformations. We have further explored the subtle physics of solutions of non-concatenated ring polymers as a model for interphase nuclei. Our results indicate that not only the territorial confinement but also the fractal, large scale crumpled loop structure arise as a generic consequence of the polymeric nature of chromosomes. Using a multi-scale approach, we are able to map the rings system onto a melt of interacting lattice animals and to identify the physics underlying the emergent behavior. Our description allows for a direct mapping to experimental data and combines massive Molecular Dynamics simulations on the fiber level with Monte Carlo simulations of a lattice model of interacting, randomly branched primitive chains for the lattice animal model. Integrated with biological information on intra- and inter-chromosomal interactions, our results pave the way for the systematic modeling of the large-scale nuclear structure and dynamics.
Monday December 2, 2013, 15:00, Room A1.04 (Science Park)
Speaker: Prof. Dr. Eric Vanden-Eijnden
Courant Institute at the New York University, USA
Host: Peter Bolhuis
Title:
Simple models for complex protein dynamics
Abstract:
The prevailing framework for the description of activated processes such as conformational changes in macromolecules is to characterize them as the hopping over a free energy barrier associated with the motion of the system along a specific reaction coordinate. Indeed this is the picture underlying classical tools such as transition state theory or Kramers' reaction rate theory, and it has been successful in explaining activated processes in a wide variety of contexts. However, there is mounting experimental evidence that this two-state picture is too simplistic to describe complex biochemical processes such as protein folding, enzyme kinetics, or protein-protein interactions. In this talk, I will review recent data from single molecule experiments and present a series of models with increasing complexity to explain the experimental findings and elucidate their microscopic origin. These results suggest that the wide range of timescales in the internal protein dynamics profoundly impact their folding pathways and rates.
Monday November 11, 2013, 10:00, Room C1.112 (Science Park)
Speaker: Prof. Dr. Ignacio Pagonabarraga
University of Barcelona, Spain
Host: Evert-Jan Meijer
Title:
Emergent structures in the collective motion of molecular motors
Abstract:
Active systems generate motion due to energy consumption, usually associated to their internal metabolism or to appropriate, localized chemical reactivity. As a result, these systems are intrinsically out of equilibrium and their collective properties result as a balance between their direct interactions and the indirect coupling to the medium in which they displace. Therefore, a dynamical approach is required to analyze their evolution and quantify their self-assembly and ability to generate intermediate and large scale stable structures. I will analyze the peculiarities of collective transport of molecular motors along biofilaments, and the patterns they give rise to due to their coupling to the embedding fluid medium in which they displace. I will also analyze the impact that confinement has in the effective motion of molecular motors and the possibility that confinement modifies the intrinsic molecular motor rectification.
Wednesday February 27, 2013, 11:00, Room D1.162 (Science Park)
Speaker: Dr. Jacco van de Streek
Department of Pharmacy, University of Copenhagen, Denmark
Host: Aurora J. Cruz Cabeza
Title:
Dispersion-corrected DFT and Tailor-made Force Fields for Materials Science
Abstract:
The molecular solid state can be modelled very reliably, in terms of structure as well as energy, with Density Functional Theory (DFT) methods, provided that a dispersion correction is added. The disadvantage of quantum-mechanical methods such as DFT is that they can only be applied to model small systems and small time scales. Force fields can be used to model much larger systems and much larger time scales, but they are not as accurate. A possible solution is the use of tailor-made force fields: force fields that are tailor-made for each molecule of interest. By dropping the condition of transferability between different chemical compounds, more parameters can be adjusted and the force fields can be of much higher accuracy. The presentation will briefly describe dispersion-corrected DFT and its performance for molecular crystal structures, followed by the description of an automated system to parameterise tailor-made force fields for chemical compounds against dispersion-corrected DFT. The last part of the presentation will outline future work, in which tailor-made force fields will be combined with Molecular Dynamics or Monte Carlo simulations.
Wednesday March 20, 2013, 14:00, Room D1.110 (Science Park)
Speaker: Dr. Marc Joyeux
Laboratoire Interdisciplinaire de Physique, Université Joseph Fourier, Grenoble, France
Host: Jocelyne Vreede
Title:
Coarse grain modelling of DNA/protein interactions
Abstract:
In this talk, I will present recent results dealing with facilitated diffusion and H-NS mediated compaction of bacterial DNA, which were obtained by investigating the properties of coarse-grained models. These models, where 15 DNA base pairs are represented by a single site and proteins by a few ones, lack most of the details of atomistic models and are not appropriate for investigating specific interactions between precise DNA sequences and proteins. In contrast, they are sufficient to model certain non-specific (mostly electrostatic) DNA/protein interactions and allow for the numerical integration of quite long trajectories for rather large systems.
Facilitated diffusion describes the alternation of 3D diffusion in the medium and 1D sliding along the DNA, by which some proteins, like transcription factors, search for their target in a DNA sequence. The question, which I will address in the first part of the talk, is whether coarse-grained models agree with or cast doubts on the popular belief according to which facilitated diffusion can accelerate the search speed by order(s) of magnitude.
H-NS is a small nucleoid-associated protein, which is involved in both gene regulation and bacterial DNA compaction. Despite the enormous effort of many research groups, there are several aspects of H-NS and H-NS/DNA interactions, which remain controversial or unknown. In the second part of the talk, I will discuss the light shed by coarse-grained models on the antagonistic effects on DNA compaction of H-NS binding to DNA in trans and cis configurations and on the possible consequences of the deposition of H-NS/DNA complexes on a surface.
Friday January 18, 2013, 11:00, Room C0.110 (Science Park)
Speaker: Prof. Dr. Paolo Carloni
German Research School for Simulation Sciences
Jülich, Germany
Host: Evert-Jan Meijer
Title:
Computational Molecular Medicine
Abstract:
The dauntingly complex functioning of human cells is often the outcome of several molecular processes. Understanding such processes is crucial for modern drug discovery, defining interaction cascades, assessing the effects of mutations changes in local concentrations of ligands, and so forth. Computational methods, from systems biology to bioinformatics and molecular simulation, allow to access features difficult or impossible to be measured. Models (if properly validated against experimental data) help understand the intricate molecular mechanisms of life processes. Bolstering the predictive power of these models calls upon the computational biologist to improve algorithms and methods. This talk will report on procedures and on applications facing current challenges in computational biology.
Thursday October 25, 14:00, Room D1.113 (Science Park)
Speaker: Dr. Marco De Vivo
Department of Drug Discovery and Development
Italian Institute of Technology, Genoa, Italy
Host: Bernd Ensing
Title:
Structure-based identification of multitarget FAAH/COXs inhibitors: tackling inflammation with a single molecule acting on multiple proteins
Abstract:
The modulation with a single compound of diverse proteins involved in a complex disease represents one of the frontiers of drug discovery programs. Here, I report on the rational structure-based identification of multitarget inhibitors that simultaneously inhibit the fatty acid amide hydrolase (FAAH) and both cyclooxygenases (COX1 and COX2) proteins. The concomitant inhibition of those proteins has been recently shown to produce in vivo improved efficacy to treat inflammation and pain - major therapeutic areas for drug discovery - with fewer side effects. First, I will show classical molecular dynamics (MD) simulations of FAAH in complex with its main endogenous substrate anandamide, or relevant covalent inhibitors used to generate clinical candidates. These studies enrich the interpretation of existing structural data and help in the clarification of how FAAH induces catalytically significant conformational states to form the acyl-enzyme adduct with the substrate, or inhibitor. These insights on the catalytic mechanism may also help in understanding other lipid hydrolase enzymes, as well. Then, I will present a multidisciplinary effort involving computational and medicinal chemistry, crystallography and pharmacology, to identify and characterize multitarget inhibitors showing promising IC50 values versus the three targets, FAAH, COX1 and COX-2. These new compounds are the most potent multitarget FAAH/COXs inhibitors reported so far in the literature and thus may represent promising starting points for the discovery of new analgesic and anti-inflammatory drugs.
Friday December 16, 14:00, Room C1.113 (Science Park)
Speaker: Prof. dr. R. Evans
Physics Department
University of Bristol
Host: Peter Bolhuis
Title:
Phase Behaviour and Structure of Confined Fluids: a Novel Delocalized Interface Phase.
Abstract:
Confined fluids (or Ising magnets) exhibit phase behaviour different from and richer than the bulk behaviour. We consider a simple fluid confined between two parallel walls (substrates), separated by a distance L. The walls are either (i) identical or (ii) exert different surface fields so that one wall is attractive and is wet by liquid (solvophilic) while the other is repulsive and may be dry, i.e. wet by gas(solvophobic). Case (i) is well-studied and we shall summarize what is known. In case (ii) a 'delocalized interface phase' forms in the temperature range Tcb >T> Tw, where Tcb and Tw are the bulk critical and wetting transition temperatures, respectively. This phase exhibits a liquid-gas interface near L/2 with pronounced thermal fluctuations. Its properties are investigated using an effective interfacial Hamiltonian approach and a fully microscopic classical density functional theory. For the physically relevant case of a fluid with r-6 interatomic attraction, we determine the scaling functions of the solvation force (the excess pressure due to confinement), the adsorption and the correlation length of interfacial fluctuations for temperatures removed from Tcb. We argue that these results remain valid beyond the mean-field treatment. In this temperature range, and for large separations, the solvation force is repulsive and decays as L-3. This result is compared with the so-called critical Casimir force which pertains for T near Tcb.
About the speaker:Prof Evans is a member of the Theoretical Physics Group in the University of Bristol. His research interests lie in the statistical physics of bulk liquids, interfacial phenomena and fluids in confining geometries. The emphasis is on understanding equilibrium structure, phase behaviour, critical phenomena, nucleation and certain dynamical properties of liquids from a microscopic viewpoint. Systems which are currently investigated include simple (atomic) fluids, ionic liquids and complex fluids, especially colloidal suspensions.
December 7, 14:00, Room C0.110 (Science Park)
Speaker: Prof. David Chandler
University of California, Berkeley
Berkeley, California, USA
Host: Peter Bolhuis
Title:
Principles of self-assembly, where kinetics can trump thermodynamics
Abstract:
Self assembly refers to the emergence of robust uniform nano-scale structures from the dynamics of a random mixture of elementary building blocks. With examples that include micellization, protein-protein interactions and virus capsid formation, three aspects of self assembly are considered in this lecture: 1. hydrophobic effects, which relate to the most common and universal forces of nano-scale self assembly; 2. competition between forces of assembly and forces that control sizes of assembling structures; 3. dynamical consequences of this necessary competition, most importantly that kinetics dominates thermodynamics.
About the speaker:David Chandler is currently Bruce Mahan Professor of Chemistry at the University of California, Berkeley. He received his S.B. degree in Chemistry from MIT in 1966, and his Ph.D. in Chemical Physics at Harvard in 1969. He began his academic career as an Assistant Professor in 1970 at the Urbana-Champaign campus of the University of Illinois, and became a full Professor in 1977. Prior to joining the Berkeley faculty in 1986, Chandler spent two years as Professor of Chemistry at the University of Pennsylvania. Chandler's primary area of research is statistical mechanics, which he has used to create many of the basic techniques for understanding condensed matter chemical equilibrium and chemical dynamics with molecular theory. He provided the modern language and concepts for describing structure and dynamics of polyatomic liquids, a series of contributions that has allowed quantitative and analytical treatments of simple molecular fluids, of aqueous solutions and hydrophobic effects, and of polymeric melts and blends. He has also developed the methods by which rare but important events can be simulated on computers, techniques that have culminated in the development of a statistical mechanics of trajectory space. This work has enabled studies of systems far from equilibrium, including processes of self-assembly and the glass transition.
December 2, 14:00, B1.25 (Beta Lounge, Science Park)
Speaker: prof. dr. T. Woolf
Johns Hopkins University, School of Medicine
Baltimore, MD, USA
Host: Peter Bolhuis
Title:
Towards Determining Order Parameters for Protein Conformational Change
About the speaker:
Tom Woolf is Professor in the Department of Physiology at he School of Medicine of Johns Hopkins University. He specializes in computer simulations of membrane proteins.
Wednesday July 6, 14:00, Room D1.113 (Science Park)
Speaker: Dr. Jed Pitera
IBM Almaden Research Center
San Jose, California, USA
Host: Bernd Ensing
Title:
Computational chemistry at IBM Almaden
Abstract:
Our research group at the IBM Almaden Research Center in San Jose, CA USA makes use of a broad spectrum of computational tools and supercomputer resources to model and understand nanoscale phenomena in soft materials. Our tools range from traditional ab initio quantum chemistry through atomistic simulation to particle- or field-based mesoscale models. In this talk, I will present examples from our current research programs in protein folding, polymer membranes for water desalination, and nanostructured antibacterial polymers. In each of these areas, we use a mix of analytic theory, coarse-grained simulation, and atomistic modeling to understand fundamental phenomena, give insight into experiments, and aid the design of new materials. In the area of protein folding, a collaborative effort with Martin Gruebele at UIUC used insights from large-scale thermodynamic and kinetic simulations as well as a computational screening protocol to shrink the 80 amino acid lambda repressor fragment to 58 amino acids without disrupting its folding thermodynamics or kinetics. In the antibacterial polymer project, I will describe how a mixed coarse-grained and atomistic simulation model helped us to understand how a change of just two methylene units produces a radical change in the supramolecular morphology of an antibacterial polymer.
Wednesday June 15, 14:00, Room D1.115 (Science Park)
Speaker: Dr. Titus van Erp
Centre for Surface Chemistry and Catalysis, Catholic University of Leuven, Belgium
Host: Peter Bolhuis
Title:
Theoretical approaches for studying zeolite formation
Abstract:
Gaining insight in the mechanism of zeolite formation has not only fundamental scientific importance, but could also lead to momentous technological developments. The applications of zeolites are uncountable ranging from gas separation, detergent builders, and sensors for pharmaceutical formulations, cracking catalysis of crude oil, and many other types of liquid- and gas-phase heterogeneous catalysis. The specific catalytic properties of zeolites lie in their unique open crystalline structure that incorporates cages or channels with typically nanoscale diameters. Theoretically, there are an uncountable number of zeolite topologies possible, but around 200 structures have been synthesized today. Despite a significant number experimental studies, a deep understanding of how zeolites form is lacking. Many models have been proposed, such as the Leuven-nanoslab model, but the models are often conflicting and, therefore, heavily debated. One of the difficulties is that the relevant length scales of the zeolite formation lie just in between the accessible length scale of NMR and diffraction techniques. This is therefore a prototype of problem where theoretical models and atomistic simulations can be complementary to experimental research.
I will present a hierarchical approach for the theoretical study of zeolites. This approach will treat the elementary steps of silicate oligomerization till the formation of the final zeolite at different levels of theory. In specific, I will show the macroscopic mathematical model for the assembly of silica nanoparticles that we derived which can explain some intriguing experimental observations. In addition, I will present some innovative methodology that I want to apply for the initial steps of silica oligomerization which is aims to combine Ab Initio MD and classical MD using path sampling (Replica Exchange Transition Interface Sampling). This approach can be considered as the dynamical equivalent of the QM/MM methodology and allows to "match" classical and Ab Initio trajectories at the barrier crossing event. On a more general level, the approach might prove to be fundamental tool for the developments of classical reactive forcefields in the future.
Thursday June 7, 14:00, Room A1.04 (Science Park)
Speaker: Dr. Luca Ghiringhelli
Fritz Haber Institute of the Max Planck Society, Berlin, Germany
Host: Evert Jan Meijer
Title:
Free gold clusters: beyond the static, mono-structure description
Abstract:
The thermodynamical stability of free, pristine gold clusters at finite temperature, and of cluster+ligands complexes at finite temperature and in the presence of an atmosphere composed of O2 and CO, is studied employing parallel tempering and ab initio atomistic thermodynamics. We focus on Au13, which displays a significant fluxional behavior: Even at low temperature (100 K) this cluster exhibits a multitude of structures that dynamically transform into each other. At finite temperature, the preference of this cluster for three-dimensional versus planar structures is found to result from entropic effects. For gold clusters containing from one to four gold atoms in an O2 + CO atmosphere, we apply ab initio atomistic thermodynamics. On the basis of these considerations, we single out a likely reaction path for CO oxidation catalyzed by gold clusters.
Thursday April 14, 11:00, Room C0.110 (Science Park)
Speaker: Dr. Erica Eiser
Cambridge University, UK
Host: Evert Jan Meijer
Title:
Flying Carpets: Colloidal Membranes via DNA driven self-assembly
Abstract:
DNA plays a special role in polymer science not just because of the highly selective recognition of complementary single DNA strands but also because natural DNA chains can be made very long, yet perfectly monodisperse. Solutions of such long DNA chains are widely used as model systems in polymer science. Here, I will present our results on the unusual self-assembly that takes place in systems of colloids coated with very long double-stranded DNA. We find that colloids coated with such long DNA can assemble into unique "floating" crystalline monolayers that are suspended at a distance of several colloidal diameters above a weakly adsorbing substrate. The formation of these 2D crystals does not depend on DNA hybridization. These crystals can be lifted and do stay stable in the bulk. Hence they have potentially interesting applications as such ordered structures can be assembled in one location and then deposited somewhere else. This would open the way to the assembly of multi-component, layered colloidal crystals.
Wednesday 9 March 2011, 14.00, Room A1.04 (Science Park)
Speaker: Dr. Aurora Cruz-Cabeza
Pfizer Institute for Pharmaceutical Materials Science, and the Cambridge Crystallographic Data Centre,
Cambridge, UK
Host: Peter Bolhuis
Title:
From Energy Landscapes and Structural Databases to Materials Design
Abstract:
This presentation will highlight the use of crystal structure prediction calculations and database analysis for the anticipation, understanding and discovery of new solid forms of pharmaceutical materials. Computer simulations and database analysis can be used to predict solid state phenomena such as polymorphism, solvate formation or the observation of inclusion compounds. The computer generation of crystal structure energy landscapes can provide new insights in the potential behaviour of organic materials and lead to the rational design of new solid forms. We are still, however, very limited on the type of systems that can be reliably predicted by the current computational methodologies. Another key aspect of the field is, therefore, to learn from current structural information found in databases and apply that knowledge into the design of more complex organic materials.
Wednesday 16 February 2011, 14.00, Room A1.04 (Science Park)
Speaker: Dr. Tom de Greef
Department of Biomedical Engineering, TU Eindhoven
Host: Peter Bolhuis
Title:
The Ins and Outs of Supramolecular Polymerization
Abstract:
The last two decades have seen a significant growth in the field of supramolecular polymerization, in which reversible, noncovalent interactions between synthetic (macro)monomeric units are utilized to build functional mesoscale architectures such as nanofibers and nanotubes. The rational design of new mesoscale architectures is hampered by the lack of a clear understanding of the self-assembly pathways that control the growth of these structures at the mesoscale. Within this talk I will derive a general classification of the growth mechanisms by which synthetic monomers self-assemble into supramolecular polymers and point to the similarities between the aggregation of proteins and synthetic monomers. Using kinetic and thermodynamic models developed to describe protein aggregation I will analyze the growth mechanism of several experimental systems. Finally, some recent efforts to understand the growth of hydrogen-bonded helical supramolecular polymers using plane-wave DFT are also discussed.
References:
1) de Greef, T. F. A.; Meijer, E. W. Nature 2008, 453, 171.
2) de Greef, T. F. A.; Smulders, M. M. J.; Wolffs, M.; Schenning, A. P. H. J.; Sijbesma, R. P.; Meijer, E. W. Chem. Rev. 2009, 109, 5687.
Tuesday 21 December 2010, 14.00, Room D1.112 (Science Park)
Speaker: Dr. Sean Sun
Johns Hopkins University, Baltimore, USA
Host: Jocelyne Vreede
Title:
Mechanochemical Processes in Biology
Abstract:
One of the goals of biology is to understand the role of genes in determining biological function, and for quantitative minded scientists, systems modeling of gene networks and ensembles has been pursued. However, a deeper look shows that the usual systems biology approach appears to be insufficient. Gene products form intricate and hierarchical microstructures whose components are highly dynamic. The microstructures can deform and exert forces; and forces, in turn, can dramatically effect enzymatic processes in microstructures. The combined effects of mechanics and chemistry are important in a number of cellular processes. Mechanical forces also naturally couple multiple cellular components, generating cooperative and nonlinear phenomena. In this talk, I will use these themes to discuss biological processes such as cell adhesion, mechanosensation and cell morphogenesis. Simple models are used whenever possible to highlight the phenomena.
Wednesday 10 November 2010, 14.00, Room C1.112
Speaker: Prof. Steven D. Schwartz
Depts. of Biophysics and Biochemistry, Albert Einstein College of Medicine, New York
and L'Institut des hautes etudes scientifiques, Bures-sur-Yvette, France.
Host: Peter Bolhuis
Title:
Reaction coordinates for complex chemical transformations - how enzymes really work
Abstract:
This talk will focus on the use of transition path sampling with analysis of the stochastic separatrix to understand the mechanism of catalysis in enzymatic reactions. We find that evolution has crafted certain enzymes to function as a "vibrational lens" that focuses heat energy in a particular direction. This motion, or promoting vibration can be used to bring reactants into close proximity, or to polarize bonds that will break. This transfer of thermal energy can by followed both theoretically and possibly experimentally. New methods to identify reaction coordinates in complex systems have been developed, and will be described.
Wednesday 7 July 2010, 14.00, Room B4.32
Speaker: Simon Poblete
Max Planck Institute for Polymer Science
Polymer Theory Group
Mainz, Germany
Host: Peter Bolhuis
Title:
Thermodynamic Forces in Adaptive Resolution Simulations.
Abstract:
-
Wednesday 23 June 2010, 14.00, Room B4.32
Speaker: Tatiana Domratcheva
Max Planck Institute for Medical Research
Department of Biomolecular Mechanisms
Heidelberg, Germany
Host: Jocelyne Vreede
Title:
Photoactivation of the novel biological blue-light receptor BLUF : a theoretical study.
Abstract:
The recently discovered blue light sensor BLUF (blue light sensing using FAD) is a flavin-based photoreceptor in Bacteria and Algae with a distinct 3D-structure that has no relation to any other known sensor protein. Despite extensive biochemical, structural and spectroscopic characterization, the mechanism of light activation and signal transduction in BLUF photosensors is poorly understood and has been under hot debate. It was demonstrated that photoactivation of BLUF involves electron transfer to the electronically excited flavin. However, the formed photoproduct state (a putative signaling state) contains a chemically unmodified oxidized flavin which is characterized by a slightly red-shifted absorption with respect to the dark adapted state. The light-induced spectroscopic shifts in BLUF were assigned to hydrogen-bond rearrangements in the chromophore binding pocket. The X-ray crystallographic studies revealed two different conformations of the receptor which were related to the dark adapted and signaling states [1].
Our efforts are aimed at identifying the photoreaction activating the BLUF light sensor by using quantum-chemical calculations. We established a computational protocol, using the CASSCF methodology, which allows a balanced description of the electronic states of different characters occurring at the BLUF chromophore site. We computed the flavin excitation energies and flavin vibrational frequencies providing the spectroscopic signatures of the dark and photoproduct states for the different BLUF models. The results suggest that the photoproduct state of BLUF contains a tautomeric form of the conserved glutamine [2]. To obtain theoretical evidence for glutamine tautomerization mediated by the electronically excited flavin, we determined the complete photoreaction pathway connecting the Franck-Condon region with the photoproduct minimum. The derived mechanism of photoactivation in BLUF is proton-coupled electron transfer (PCET) which is one of the main photochemical and redox mechanisms identified in biomolecules.
[1] Jung, A., Reinstein, J., Domratcheva, T., Shoeman, R. L.,
Schlichting, I. J. Mol. Biol. (2006) 362,
717-732
[2] Domratcheva, T., Grigorenko B. L., Schlichting I.,
Nemukhin A. V. Biophys. J.
(2008) 94, 3872 - 3879
Thursday 15 April 2010, 11.00, Room B4.32
Speaker: Serena Donnini
Max Planck Institute for Biophysical Chemistry
Department of Theoretical and Computational Biophysics
Göttingen, Germany
Host: Jocelyne Vreede
Title:
In silico titration of biomolecules: molecular dynamics at constant pH.
Abstract:
pH is an important parameter in condensed phase systems as it determines the protonation state of ionizable groups and consequently influences the structure, dynamics and function of molecules in solution. In most force field simulation protocols, however, the protonation state of a system (rather than its pH) is kept fix and cannot adapt to changes of the local environment. We present here a method to perform molecular dynamics (MD) simulations in explicit solvent at constant pH. During the simulation the protonation states of titrating groups are allowed to change dynamically reproducing average protonation probabilities at a certain pH. The method is based on the lambda-dynamics approach, where the dynamics of the titration coordinate lambda is driven by the energy gradient between the protonated and deprotonated states. Hydration free energies for deprotonation are included by empirically altering the energy landscape. By accounting for the pH dependence of the hydration free energy in the energy landscape correction, constant pH simulations can be achieved. The method is applied to systems with single or multiple titrating sites in explicit solvent and predicted pKas of aminoacid analogues and of a small peptide.
Wednesday 24 March 2010, 14.00, Room B4.32
Speaker: Nikos L. Doltsinis
Department of Physics, King's College London, London WC2R 2LS, United Kingdom
Host: Peter Bolhuis
Title:
Multiscale Modelling of Photoactive Materials
Abstract:
A DFT-based nonadiabatic ab initio molecular dynamics (na-AIMD) surface hopping scheme [1,2] has been applied to study light-induced processes in a variety of complex condensed phase systems whose relevance range from biology [3] to materials science [4,5]. As part of a long-term strategy to model macroscopic phototriggered phenomena, a multiscale simulation approach is being developed linking different hierarchical simulation levels - from na-AIMD [1,2] to atomistic classical MD to coarse grained MD. In this scheme, the develoment of a nonadiabatic QM/MM extension [4,5] of na-AIMD has played a key role. Applications to a photoswitchable liquid crystal [4,5], based on the azobenzene chromophore, and the light-triggered unfolding of a polypeptide [6] are presented.
- N. L. Doltsinis and D. Marx, Phys. Rev. Lett. 88, 166402 (2002); J. Theor. Comp. Chem. 1, 319 (2002).
- N. L. Doltsinis in Computational Nanoscience: Do it yourself!, edited by S. Blügel, J. Grotendorst and D. Marx (NIC, FZ Jülich, 2006), www.fz-juelich.de/nic-series/volume31/doltsinis1.pdf
- N. L. Doltsinis, P. R. L. Markwick, H. Nieber, and H. Langer, in Radiation Induced Molecular Phenomena in Nucleic Acid, edited by M. K. Shukla and J. Leszczynski (Springer, Netherlands, 2008), and references therein.
- M. Böckmann, N. L. Doltsinis and D. Marx, Phys. Rev. E 78, 036101 (2008).
- M. Böckmann, N. L. Doltsinis and D. Marx, J. Phys. Chem. A 114, 745 (2010).
- H. Nieber, A. Hellweg, and N. L. Doltsinis, J. Am. Chem. Soc., 132, 1778 (2010).
Wednesday 3 March 2010, 14.00, Room B4.32
Speaker: Dr. Stuart C. Althorpe
Department of Chemistry, University of Cambrige,
United Kingdom
Host: Peter Bolhuis
Title:
Path-integral rate-theory in the deep-tunnelling regime
Abstract:
---
Wednesday 7 October 2009, 14.00, Room B4.32
Speaker: Pietro Vidossich
Computer Simulation & Modeling Laboratory (CoSMoLab) in Barcelona, Spain
Host: Bernd Ensing
Title:
Heme hydroperoxidase catalysis: the role of distal side residues
Abstract:
Heme enzymes catalyze a wide range of chemical reactions and research aimed to grasp the origin of this functional diversity is an extremely active field. We present results from modeling studies of peroxidase, catalase and catalase-peroxidase from which the role played by distal side residues clearly emerges. By means of hybrid QM/MM calculations we have investigated the atomic and electronic reorganization taking place during i) the formation of the high valent iron-oxo species known as Compound I (Cpd I) in Horseradish peroxidase (HRP) from reaction with H2O2 [1]; ii) the reduction of Cpd I by H2O2 in Helicobacter pylori heme-b catalase (HPC) and Penicillium vitale heme-d catalase (PVC) [2]. In both studies we focussed on the process from the inner-sphere enzyme:H2O2 complex to the products.
Our calculations suggest that the H-bond network at the distal side plays a key role in positioning the peroxide. To facilitate the reaction, short H-bonds have to be attained and suitably positioned acid-base catalysts are needed to shuttle protons. Neither of the two reactions (i and ii above) proceed in a single step, and to achieve these requirements reorganization of the intervening species is needed, revealing the motion of catalytic residues. Protons may be transferred together with an electron, in an H atom transfer or a proton coupled electron transfer PCET. From the electronic point of view, the catalase reaction (ii above) proceeds by two one-electron transfers rather than a two-electron step, as has long been the lore.
Furthermore, we have characterized the electronic structure of Burkholderia pseudomallei catalase-peroxidase (KatG) Cpd I [3], showing that the radical character of Cpd I is affected by the protonation state of the KatG specific Met-Tyr-Trp covalent adduct present at the active site. These findings may provide a clue to understanding how catalatic activity is introduced into a peroxidatic core.
- E. Derat, S. Shaik, C. Rovira, P. Vidossich, M. Alfonso-Prieto. J. Am. Chem. Soc. 129, 6346-6347 (2007)
- M. Alfonso-Prieto, X. Biarnes, P. Vidossich, C. Rovira. J. Am. Chem. Soc. 131, 11751-11761 (2009)
- P. Vidossich, M. Alfonso-Prieto, X. Carpena, P. C. Loewen, I. Fita, C. Rovira, J. Am. Chem. Soc. 129, 13436-13446 (2007)
Wednesday 21 October 2009, 14.00, Room B432
Speaker: Dr. Richard Sear
University of Surrey, UK
Host: Peter Bolhuis
Title:
Nucleation of crystals: Experimental studies of protein crystallisation and computer simulations of simple models
Abstract:
TBA
Friday 4 September 2009, 13.00, Room B2.44
Speaker: Dr. Vanessa Ortiz
University Wisconsin-Madison, USA
Host: Bernd Ensing
Title:
Molecular Origins of DNA Flexibility: Sequence Effects on Conformational and Mechanical Properties
Abstract:
TBA
Wednesday 13 May 2009, 15.00, Room C206
Speaker: Prof. dr. David van der Spoel
Uppsala University, Uppsala, Sweden
Host: Peter Bolhuis
Title:
Kinetics and Thermodynamics from MD Simulations
Abstract:
Algorithms for deducing kinetic properties from MD simulations will be presented with applications in hydrogen-bonding and protein folding. Important properties like the Gibbs energy of activation for breaking a hydrogen bond in different environment can be deduced from a relatively simple analysis of the presence of particular hydrogen bonds as a function of time. The same sort of analysis can be applied to other processes that can be described by a few discrete states, and where multiple transitions between the states are found in the simulations.
Friday 11 December 2008, 15.00, Room B432
Speaker: Remus T. Dame
Cell Observatory, Leiden Institute of Chemistry, Leiden University and
Physics of Complex Systems, Department of Physics and Astronomy, VU University, Amsterdam
Host: Jocelyne Vreede
Title:
Unravelling the organization of bacterial chromatin by H-NS
Abstract:
The genome of bacteria is folded and compacted into a body referred to as the nucleoid due to the action of nucleoid-associated proteins (NAP's). As a consequence of their role in global genome organization these proteins also act as pleiotropic regulators of transcription. One of the key players in these processes is H-NS. This is an abundant, multimeric protein with a binding preference for A/T rich regions along the genome. Its binding to these regions is associated with transcriptional silencing and has been suggested to be a mechanism to specifically target and silence newly acquired foreign DNA and to protect the host against its potentially harmful effects.
A lot of progress in the understanding of H-NS action has been booked in recent years. Our main aim has been to establish the structure, kinetics, mode of binding and the role of H-NS in global genome organization. To this purpose we used a combination of scanning force microscopy imaging, single-molecule micromanipulation and theoretical modeling of H-NS-DNA complexes. Initially, we demonstrated that H-NS organizes DNA by bridging two DNA duplexes and put forward evidence that this mode of binding is also key to the role of H-NS as repressor. In recent follow up studies, we showed that H-NS acts as a dimer, that H-NS dimers are stacked side-by-side between bridged duplexes and determined the dimensions of the dimer. This provides clues to the configuration of the coiled-coil responsible for dimerization Moreover, we determined the binding kinetics of individual H-NS dimers and the forces required to open up H-NS-bridged regions. Finally, combining the in vitro structural observations and in vivo ChIP-on-chip data for binding of H-NS along the genome, we can now explain the higher order organization of the genome in the long known topologically isolated domains.
Friday 7 November 2008, 11.00, Room B432
Speaker: Anna Pavlova
Royal Insitute for Technology, Stockholm, Sweden / University of Amsterdam
Host: Evert Jan Meijer
Title:
Theoretical Investigation of Reductive Silylation of Imines with Trichlorosilane
Abstract:
Reduction of prochiral imines with chiral catalysts is the main pathway for obtaining chiral amines that are employed in several areas of industrial chemistry. One of these pathways is reduction of imines with trichlorosilane, where the first and also the enantioselective step is the reductive silylation of the imine. This step was investigated theoretically with Density Functional Theory calculations and the results are summarized in this thesis.
The B3LYP density functional was used in the computations. The imine used for the calculations was (N)-benzylidene-aniline which reacts with trichlorosilane in acetonitrile without addition of a catalyst. Based on the available experimental data, the reaction was assumed to initiate with the formation of a hypervalent silicon complex followed by a hydride transfer to the carbon of the imine bond.
Various pathways with both intra- and intermolecular hydride transfers were explored. It was concluded that the hydride transfer step is not intramolecular since the computed barriers were unfeasible in all of the possible cases. The calculations also strongly suggested that a mechanism involving an intermolecular hydride transfer requires a stabilization of the imine nitrogen.
Wednesday 5 November 2008, 11.00, Room C210
Speaker: Prof. Wenchuan Wang
Beijing University of Chemical Technology
Beijing, China
Host: Evert Jan Meijer
Title:
Hydrogen storage in novel materials of silicon nanotube and covalent organic frameworks (COFs): A multiscale method in computer simulation
Abstract:
TBA
Friday 25 July 2008, 11.00, Room B432
Speaker: Giovanna Zilibotti
Cecam
Lyon, France
Host: Evert Jan Meijer
Title:
Ab initio analysis of friction between C-diamond surfaces
Abstract:
The frictional properties of two surfaces in relative motion are governed, in absence of wear, by the features of the potential energy surface (PES) which describes the variation of the surface mutual interaction during the relative displacement. The static friction force of commensurate systems is completely determined by the gradient of the PES profile along the direction of motion. The kinetic friction originates from elastic instabilities: the accumulated potential energy, during the climbing of the PES saddle points, is suddenly released giving rise to excitations like phonons. First principle calculations, based on density functional theory, which combine electronic self consistency and efficient structural optimization, offer the possibility to accurately determine the PES of a sliding interface. The PES can, in fact, be constructed by calculating the interface energy (or the surface adhesion) on a grid of points corresponding to different relative positions of the two surfaces. The knowledge of the PES can provide useful information for understanding the tribological properties of nanoscale materials. Furthermore the understanding of the atomistic interactions which determine the PES maxima and minima can provide indications on how the PES corrugation can be modified in order to control friction. The systems we have studied are diamond surfaces in presence (or in absence) of adsorbates. These kind of systems are receiving attention in the field of natribology as promising materials for micro- nanoelectromechanical systems (MEMS/NEMS) and for practical applications. The adsorbates we have considered are hydrogen and OH groups, whose presence on the surfaces can be really determined by their exposition at humid air. Our analysis is centerd on three aspects:
- the role of the adsorbates (hydrogen atoms and OH groups) in surfaces stability and in friction between commensurated surfaces in relative motion;
- the role of C- dangling bonds at the interface between diamond surfaces in relative motion;
- the study of friction under load: a constant external load had been imposed on the system to compress it. Is the proportionality between load and friction force still valid on nanoscale systems?
John van Geuns Lecture
Wednesday 2 July 2008, 11.00, Room B2.44Speaker: Steven Nielsen
Department of Chemistry, University of Texas at Dallas
Dallas, TX, USA
Steve Nielsen website
Host: Bernd Ensing
Title:
Nanoparticle localization energy at the oil/water interface: How to fix Young's equation
Abstract:
Young's and the modified Young's equation stand as the current theory of nanoparticle localization energy at the oil/water interface. These equations relate the interaction energies between the solid, oil, and water phases to the contact angle at which the three phases meet. It is well-known, however, that these theories fail to describe experimental or computer simulation data. Here we show, using free energy molecular dynamics calculations, that accounting for the curvature dependence of surface tension leads to a correct theory. In addition, we analyse the role of line tension. In the context of the modified Young's equation, we show that line tension depends on the nanoparticle radius. Upon further analysis, however, we argue that the apparent line tension in the modified Young's equation arises from an incorrect assignment of nanoparticle radius in Young's equation. The apparent line tension is negative if the radius is underestimated, and positive if the radius is overestimated. We conclude that the modified Young's equation is not needed, and suggest that we have resolved the controversial issue of the existance and sign of the line tension.
Friday 23 May 2008, 11.00, Room B432
Speaker: Wouter den Otter
Computational Biophysics, Universiteit Twente
Enschede
Host: Peter Bolhuis
Title:
Simulations of lipid bilayers: dynamics and pores.
Abstract:
Biological membranes are self-assembling bilayers of amphiphilic molecules. Because bilayers are held together by relatively weak non-bonded interaction forces, they behave in many respects as two-dimensional liquids suspended in a three-dimensional solvent matrix. This makes bilayers very susceptible to external forces, which give rise to flow within the bilayer and to deformations of the overall shape of the bilayer, including rupture. We present molecular dynamics simulations of these processes, using both coarse-grained and atomistic models of lipid membranes.
Surface shear viscosity and intermonolayer friction have been studied in equilibrium and non-equilibrium simulations, studying their dependence on the lipid architecture. Membranes under lateral tension are employed to investigate the stability of transmembrane pores and the line tension of bilayer edges. The free energy of opening up a transmembrane pore is explored through the potential of mean constraint force method. As a final topic, we delve into the mechanisms by which dimethylsulfoxide (DMSO) enhances the permeability of our skin.
Friday 25 April 2008, 11.00, Room B432
Speaker: Oliwia Szklarczyk
Biophysics, Warsaw university
Warsaw, Poland
Host: Peter Bolhuis
Title:
Electrostatic Contributions to the Free Energy of Binding of mRNA caps to the eIF4E Translational Factor
Abstract:
Electrostatic interactions are believed to play a central role in many biological processes. Electrostatic binding free energies were calculated for six enzyme-ligand complexes. A Poisson-Boltzmann model was used with two different protocols implemented: 1) the dielectric boundary determined as the solvent exclusion surface with the protein dielectric constant ep=4 and 2) the dielectric boundary determined as the van der Waals surface with ep=4. Results of both of the models were corrected with estimated entropic contributions to the binding free energy and compared with the experimental data as well as discussed regarding previous research on the theoretical approach.
Tuesday, 15 April 2008, 14:00-15:00, Room B432
Speaker: Carla Bosia
University of Turin - Faculty of Mathematics, Physics and Natural Science
Theoretical Science group
Turin, Italy
Host: Evert Jan Meijer
Title:
Nucleation phenomena in biological systems
Abstract:
One of the most important and still not completely understood characteristics of the complex cellular activity is chemotaxis under gradient toward a chemical attractant. From a biological point of view, one of the crucial molecular mechanisms is the polarization of the cell membrane: two enzymes, phosphatidylinositol 3-kinase (PI3K) and phosphatase and tensin homolog (PTEN), inter-convert the phospholipids PIP2 and PIP3 (phosphatidylinositol bisphosphate, PIP2, and phosphatidylinositol trisphosphate, PIP3). The interplay among these enzymes leads to a process of phase separation. The final imbalance between two resulting domains causes the formation of a growing head and a retracting tail. In such a way, the cells can start to move towards the source of the chemical signal. This process has many features in common with the well studied nucleation phenomena in statistical mechanics and can be modelled with a suitable generalization of the two dimensional Ising model in a magnetic field. The goal of this thesis work is to extend the model to a cylindric geometry, in order to describe the chemotaxis process for bacilli or for philopodia motility.
7 February 2008, 14:00, Room B432
Speaker: Manuel Dömer
Universität Zürich EMPA - Materials Science & Technology ,
Zurich, Switzerland
Host: Evert Jan Meijer
Title:
Towards Two-Level Force Fields for Photochemical Switches
Abstract:
The most prominent example of a photochemical switch in biological systems is the Retinal chromophore in the active site of Rhodopsin, the visual pigment in the eyes of vertebrates. It illustrates the basic principle of such a switch in a very impressive way: The light induced E/Z-isomerization within the Retinal molecule results in a conformational change of a molecular scaffold linked to the switch. In the search for new artificial photochemical switches based on this principle computational methods are used to design and characterize potential candidates preliminary to the, often very time consuming, quest for a synthetic realization. The involvement of electronic excited states in a photochemical reaction calls for computationally very expensive quantum chemical methods, like the complete active space self consistent field (CASSCF) approach, a method unaffordable for realistic systems, like a photochemical switch embedded in a solvent or in a protein, and for the calculation of a statistically significant number of molecular dynamics trajectories. In order to get access also to dynamical properties, like the quantum yield of a photochemical reaction, via MD I investigate the possibility to represent both the electronic ground and the first excited state by a classical force field and therewith reduce greatly the expense of energy and gradient evaluations. These new force fields are based on the Generalized Amber Force Field (GAFF) with only a few additional coordinates representing the reaction coordinate of the E/Z-isomerization of the switch. The parameters of the novel force fields are based on CASSCF calculations on a suitable grid (indeed, the present work is part of a collaboration with M. Olivucci and F. Santoro (Siena and Pisa, Italy) who performed the necessary CASSCF based evaluations). Several issues concerning the consistent merging of the new terms into the existing parameter set have been faced and solved. Routines to handle input keywords, energy and gradient calculations of these new coordinates are implemented within the TINKER suite of programs.
December 7, 2007, 13:00, Room B4.32
Speaker: Dr. Arturo Robertazzi,
Dept. of Statistical and Biological Physics
Sissa
Trieste, Italie
Title:
Theoretical studies of systems of biological interest
Abstract:
Several scientific areas have benefited from the development of theoretical tools that have now reached maturity to complement experimental studies. Chemistry, biochemistry and medicinal chemistry are attractive fields that are crucial, for example, for developing therapies against those diseases that affect humanity. In this work we present several examples of how theoretical methods, from quantum mechanical calculations to molecular dynamics, can be applied to address issues of biological interest. In particular, discussion will focus on novel strategies for developing more effective anticancer drugs.
Anticancer research achieved
astonishing results in the past decades[1]. Several compounds,
often containing heavy metals, were found to be active against
cancer[2,3]. Among the severe drawbacks that might reduce the
applicability of anticancer drugs, the selectivity towards
their biological target, most often DNA, is of a major
concern[1]. A possible strategy to overcome these issues is
the combination of different drugs of known activity
(multifunctional drug approach). We are currently involved in
the investigation of intriguing examples of the latter, in
collaboration with prof. J. Reedijk and co-workers who
recently combined cisplatin-derivatives with copper
phenanthroline DNA cleavers, Figure 1 [4,5]. Since the binding
mode of platinum drugs has been extensively addressed, our
work focuses on copper phenanthroline complexes, whose DNA
binding mode is still debated. Combination of several
theoretical techniques and a careful analysis of experimental
data allowed us to address two crucial issues: i) the mode of
binding of the copper phenanthroline complexes ii) the
ligand-structure/DNA-cleavage-efficiency relationship. In
summary, this study provides an extensive description of the
copper phenanthroline/DNA adducts that is crucial for the
rational design of novel anticancer complexes in which the DNA
cleavers are combined with known drugs[5-7].
Cartoon representing the most efficient DNA cleaver
KEY REFERENCES
(1) Hurley,
L. H. Nat. Rev. Cancer 2002, 2, 188-200.
(2) Jamieson,
E. R.; Lippard, S. J. Chem. Rev. 1999, 99, 2467-2498.
(3) Alessio, E.; Mestroni, G.; Bergamo, A.; Sava,
G. Curr. Top. Med. Chem. 2004, 4, 1525-1535.
(4) de
Hoog, P.; Boldron, C.; Gamez, P.; Sliedregt-Bol, K.; Roland,
I.; Pitie, M.; Kiss, R.; Meunier, B.; Reedijk,
J. J. Med. Chem. 2007, 50, 3148-3152.
(5) Robertazzi,
A.; Magistrato, A.; deHoog, P.; Carloni, P.; Reedijk,
J. Inorg. Chem. 2007, 46, 5873-5881.
(6) Pitié, M.;
Boldron, C.; Pratviel, G. Adv. Inorg. Chem. 2006, 58, 77-130.
(7) Robertazzi, A.; Vargiu, A.; Magistrato, A.; de Hoog,
P.; Ruggerone, P.; Carloni, P.; Reedijk, J. Biophys. J. 2007,
(to be submitted).
October 9, 10:00, location will be annouced later
Speaker: Dr. Bernd Ensing,
Dept. of Chemistry and Applied Biosciences
ETH Zurich, Switzerland
Title:
Metadynamics as a Tool for Exploring Free Energy Landscapes of Chemical Reactions
Abstract:
The metadynamics or hills method is a relatively new molecular dynamics technique aimed to enhance the sampling of separated regions in phase space and map out the underlying free energy landscape as a function of a small number of order parameters or collective variables. The high efficiency allows for the application of metadynamics in combination with first principles dynamics methods, in particular with Car-Parrinello molecular dynamics, to study processes in which changes in the electronic structure play a dominant role, such as chemical reactions. The option to choose several independent collective variables is important to tackle complex and concerted transformations that lack an obvious a priori choice for a single reaction coordinate. I will discuss the role of metadynamics in the search of transition states, local minima, reaction paths, free energy profiles, and reaction coordinates among a growing list of alternative methods.
October 13, 11:00, R de Vries Room, Plantage Muidergracht, N-building
Speaker: Dr. Daniele Coslovich,
Department of Theoretical Physics
University of Trieste, Italy
Title:
LOCALIZED SADDLES OF THE POTENTIAL ENERGY SURFACE AND DYNAMICAL
HETEROGENEITIES IN SUPERCOOLED LENNARD-JONES LIQUIDS
Abstract:
Understanding the peculiar properties of dynamics in supercooled liquids remains an open and stimulating challenge in condensed matter physics. The existence of localized regions of high mobility (dynamical heterogeneities) in the liquid is one of the most intriguing features which appears upon supercooling [1]. In the last few years, several efforts have been done in order to explain dynamical heterogeneities in terms of fundamental properties of the system, but their origin remains unclear. In particular, the paradigmatic description of the glass transition in terms of the Potential Energy Surface (PES) [2] has not been able so far to account for these dynamical processes. Recently, we have suggested [3] that the properties of negatively curved regions of the PES might possess, indeed, a deep connection with dynamical heterogeneities. In this talk, I will present some results from Molecular Dynamics simulations of a supercooled Lennard-Jones mixture, showing a correlation between dynamical heterogeneities and the properties of saddles and quasisaddles of the PES. Upon cooling the liquid towards the estimated Mode-Coupling critical temperature, we observe a strong localization of unstable modes and we find that the spatial regions of the liquid where this localization occurs are correlated to clusters of mobile particles. These results point to a key role of saddles and quasisaddles of the PES in explaining the emergence of a heterogeneous dynamics in supercooled liquids.
[1] Widmer-Cooper A. et al., Phys. Rev. Lett. 93, 135701 (2004)
[2] Angelani L. et al., J. Chem. Phys. 119, 2120 (2003)
[3] Coslovich D. and Pastore G., Europhys. Lett. 75, 784 (2006)
23 June 2006
Speaker: Dr. Galina Kalibaeva,
Computational Physics Group
Universita di Roma "La Sapienza", Italy
Title:
Simulation of a granular system using hardspheres model with holonomic constraints.
Abstract:
In the first part of this presentation we demonstrate the possibility of including constraints in hard systems, using the simple case of a dimer of hard spheres, where the analytical solution exists. We make a detailed description of the model and show that the system's dynamics can be solved in a rigorous way. In the second part we slightly modify the model, allowing the inelastic collisions. We apply this model to a bidimentional gas of inelastic dimers, and obtain some caracteristics of the system, such as the decay of the total kinetic energy, and the ratio between its rotational and translational parts. We also measure the distribution of velocities in the gas as it cools.
Thursday April 27,
Speaker: Dr. Rodolphe Vuilleumier,
Laboratoire de Physique Theorique de Liquides
Universite Pierre et Marie Curie, Paris
Title:
Vibrational spectroscopy at finite temperature from ab initio Molecular Dynamics simulations
Abstract:
Vibrational spectroscopy, IR, Raman etc., is a widespread analytical tool for getting information about the structure and dynamics of chemical species. Ab initio Molecular Dynamics simulations have proved to be able to reproduce in many situations the observed spectra. We will present here some recent developments in the analysis of the simulated spectra. We will consider separately the extraction of vibrational modes from the Molecular Dynamics simulations and the analysis of the band intensities. The former are linked to the atomic polar tensors and an exact sum rule of the infrared spectrum can lead to well defined atomic charges derived from these atomic polar tensors. Two examples will be given from studies in the liquid phase at ambient temperature, both for the extraction of vibrational modes and for the analysis of the atomic polar tensors: a solvated N-methyl-acetamide molecule (peptide bond model) in water and liquid water itself.
Host: Evert Jan Meijer
Friday April 7, 11.00, B4.32
Speaker: Francesco Colonna,
Universita' di Trieste
Title:
Fluid phase diagram of binary mixtures from integral equations.
Thursday April 6, 14.00 location to be confirmed
Speaker: Karsten Reuter,
Fritz-Haber-Institut der Max-Planck-Gesellschaft, Faradayweg 4-6, D-14195 Berlin, Germany
Title:
First-Principles Statistical Mechanics for Heterogeneous Catalysis
Abstract:
First-principles electronic structure theory calculations have become an increasingly important tool for the predictive modelling of materials properties at the atomistic level. In order to reach technically relevant length and time scales a link with concepts from statistical mechanics and thermodynamics has now to be found. I will sketch corresponding multi-scale modelling attempts for the field of heterogeneous catalysis, concentrating particularly on the relevance of treating the surrounding gas phase, as well as the statistical interplay of the manifold of elementary processes at a catalyst surface.
Friday Februari 24, 11.00 B4.32
Speaker: T. Zykova-Timan,
Department of Computational Chemistry ETH-Zurich
Title:
"Why NaCl(100) does not wet by its own melt?"
Abstract:
NaCl (and other alkali halide) crystal surfaces have the peculiar property of repelling their own melt. As a result they let themselves be wetted only partially by their own liquid at the melting point Tm. We recently investigated the physical reasons for this unusual behavior. We found them through theory and molecular dynamics simulation to stem from the conspiracy of three factors. First, the solid NaCl(100) surface is exceptionally anharmonic,but also exceptionally stable. It can in fact survive even well above the melting point, for unlike most other surfaces it does not spontaneously melt. Second, the solid-liquid interface is very costly, due to a 27% density difference between solid and liquid. Third, the surface tension of liquid NaCl is relatively high. This last feature is due to an unexpected entropy deficit, that can in turn be traced to incipient molecular charge order in the outermost regions of the molten salt surface.[1]
[1] T. Zykova-Timan, D. Ceresoli, U. Tartaglino and E.Tosatti, PRL 94, 176105 (2005).
Friday January 20, 11.00, B432
Dr. Johan Carlsson
MPI Fritz Haber Insitute, Berlin
Title:
"A microscopic view into nanoporous carbon materials"
Abstract:
Carbon is one of the most versatile elements in nature and Nanoporous carbon (NPC) constitutes yet another class of carbon materials that exhibit unusual properties. NPC has the ability to catalyze dehydrogenation reactions, but the actuating chemical reaction steps are still unclear. This uncertainty is also due to the fact that the atomic structure of NPC depends on the preparation conditions. TEM experiments suggest that NPC derived from hydrocarbons has the form of crumpled graphene sheets with a significant amount of non-hexagonal rings in the structure. We have therefore carried out an extensive study to characterize NPC. Our density-functional theory (DFT) calculations reveal that the atomic relaxation transforms defects into combinations of non-hexagonal rings, which we identify as the ``motifs of NPC''. These motifs lead to strain and local buckling of the structure. They also induce defect states close to the Fermi level, leading to that some of them being charged, which may facilitate molecule dissociation. These motifs can then be combined to build models of new carbon materials. A random distribution of the motifs leads to the formation of a NPC material and our calculations indicate that this type of NPC can have a heat of formation comparable to other metastable carbon materials such as nanotubes.
Tuesday 10 May 2005 - 14.00 - B432
Prof. Sean Sun,
Whitaker Biomedical Engineering Institute,
John Hopkins University, Baltimore (USA)
Title:
Proton transfer through narrow pores.
Abstract:
The crawling movement of animal cells is a beautiful example of spatial and temporal organization in biological systems. We examine two commonly encountered organelles in cell motility, the lamellipodium and the filopodium, using coarse-grained models. The roles of the cell membrane and membrane proteins in generating and maintaining the moving cell are explored. We will show that the moving mechanism is intimately related to the properties of the cell membrane.
28 April 2005 - 15.00 - B432
Prof. dr. Christoph Dellago, Universiteit van Wenen.
Title:
Proton transfer through narrow pores.
Abstract:
One-dimensional chains of hydrogen bonded water molecules provide excellent conductors for protonic currents through pores across membranes, a process which is fundamental to many biological and technological systems. In this seminar I will report on the results of computer simulations of proton transfer along water chains inside carbon nanotubes. Ab initio molecular dynamics and an empirical valence bond model yield similar structures and time scales. We find that the proton mobility along 1D water chains exceeds that in bulk water by a factor of 40, but is reduced if orientational defects are present in the water chain. Excess protons interact with hydrogen-bonding defects through long-range electrostatics, resulting in coupled motion of protons.
1 March 2005 - 14.00 - B432
Ali Alavi (Theoretical Chemistry, Cambridge University, UK)
Title:
A Quantum Monte Carlo method for correlated Fermion systems based on sampling Slater determinants
Abstract:
The simulation of Fermion systems has long been plagued with "sign-problems", both in zero-temperature diffusion quantum MC and in T>0 path-integral MC. In this talk, I will present a method in which the sampling of discrete antisymmteric spaces - Slater determinants constructed out of given set of one-electron orbtials - is performed. Some new analytic results which allow path counting to be performed in such spaces have been derived, and these are used to construct an QMC algorithm to sample Slater determinant space. The method has been applied to strongly correlated Hubbard models and also to the problem of dissociating diatomic molecules, with promising results.
Host: Evert Jan Meijer
22 February 2005 - 14.00 - B432
Joost Vandevondele (Theoretical Chemistry, Cambridge University, UK)
Title:
A molecular dynamics study of the hydroxyl radical in solution applying
self-interaction corrected density functional methods
Abstract:
To be announced.
Host: Evert Jan Meijer