Challenges across Large-Scale Biomolecular and Polymer Simulations
February 21, 2017 - February 24, 2017 CECAM-AT
Ivan Coluzza University of Vienna, Austria
Samuela Pasquali Paris Descartes University, France
Barbara Capone University of Vienna, Austria
Christoph Dellago University of Vienna, Austria
Tamar Schlick New York University, USA
1 Description
The goal of this workshop is to bring together scientists who study large scale molecular systems both in the biological world and those from the polymer science community. This workshop aims to bring broad-minded scientists interested in these challenges to discuss state-of-the-art approaches and issues with the goal of advancing structural biophysics and applied fields. By discussing various approaches and ways to integrate knowledge on multiple scales, we hope to build a community to advance this important field. The increase in computer power in tandem with algorithmic and force-field improvements are opening opportunities for modeling large biomolecular systems as never before. Yet, many challenges in modeling and simulations of such complex systems, in realistic environments, require innovations to make the necessary experimental connections and develop new theories and mechanistic details of the associated processes. The workshop that we propose will follow and build upon the first workshop organized in Telluride (June 14-18, 2015) by Tamar Schlick and Klaus Schulten entitled "Challenges in Simulating Large-Scale Biomolecular Complexes", by focusing on the following subjects: - Large scale simulations with respect to time and system size. We will include recent advances in atomistic force-fields and technological advances in algorithms and hardware that make possible simulations on the millisecond time scales for systems composed of millions of atoms, such as large solvated biomolecules. - Quantitative coarse-grained models going toward multi scale approaches. We will highlight the development of models beyond atomistic with different resolution tailored to the problem under investigation. Such approaches are essential for addressing problems that are still out of reach by atomistic simulations, even with the best hardware and algorithms. Moreover, some collective phenomena and statistical properties of the systems emerge more naturally from a representation of the systems focusing only on the relevant degrees of freedom, approximating specific details. - Bridging the gap between polymer physics and large biomolecular systems. We aim to unite two communities where complementary approaches are developed for the study of large molecular systems both on theory, simulation algorithms and experimental techniques. Several examples already show that the overlap between these communities can provide high impact results, including chromatin and RNA modeling. All these areas will be represented by scientists with either a theoretical or experimental background. In the latter case, we will aim to bring experimentalists with modeling experience as well. All the above-mentioned subjects are crucial for the further development of large scale molecular simulations. However, scientists from such distant disciplines rarely have the chance to come together and exchange their point of view on the common problem of large molecular simulations. With this workshop, we intend to create a long-lasting discussion table that in the future will be the base of an established common effort for the design of self-assembling systems both biological and artificial.
The workshop will start the 21st of February at 14:30 and end the 24th at 12:45. The venue for the meeting is: https://goo.gl/maps/qJqQJtxgNXG2
Key references [1] T. Schlick, R. Collepardo-Guevara, L. A. Halvorsen, S. Jung, and X. Xiao, Quarterly Reviews of Biophysics 44, 191 (2011). [2] J. H. Konnert and W. A. Hendrickson, Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoretical and General Crystallography 36, 344 (1980). [3] A. Jack and M. Levitt, Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoretical and General Crystallography 34, 931 (1978). [4] A. Warshel and M. Levitt, Journal of molecular biology 103, 227 (1976). [5] J. A. McCammon, B. R. Gelin, and M. Karplus, Nature 267, 585 (1977). [6] J. C. Phillips, R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot, R. D. Skeel, L. Kalé, and K. Schulten, Scalable molecular dynamics with NAMD (Wiley Online Library, 2005), 26, p. 1781. [7] K. J. Bowers, E. Chow, H. Xu, R. O. Dror, M. P. Eastwood, B. A. Gregersen, J. L. Klepeis, I. Kolossvary, M. A. Moraes, and F. D. Sacerdoti, Proceedings of the 2006 ACM/IEEE conference on Supercomputing (ACM Tampa, FL, 2006). [8] H. J. C. Berendsen, D. van der Spoel, and R. van Drunen, Computer Physics Communications 91, 43 (1995). [9] Y. Duan, L. Wang, and P. A. Kollman, Proceedings of the National Academy of Sciences 95, 9897 (1998). [10] Y. Duan and P. A. Kollman, Science (New York, N.Y.) 282, 740 (1998). [11] M. Shirts and V. S. Pande, Computing 10, 43 (2006). [12] C. D. Snow, H. Nguyen, V. S. Pande, and M. Gruebele, nature 420, 102 (2002). [13] D. L. Ensign, P. M. Kasson, and V. S. Pande, Journal of molecular biology 374, 806 (2007). [14] B. Zagrovic, E. J. Sorin, and V. Pande, Journal of molecular biology 313, 151 (2001). [15] P. L. Freddolino, A. S. Arkhipov, S. B. Larson, A. McPherson, and K. Schulten, Structure 14, 437 (2006). [16] A. Pérez, F. J. Luque, and M. Orozco, Journal of the American Chemical Society 129, 14739 (2007). [17] P. Maragakis, K. Lindorff-Larsen, M. P. Eastwood, R. O. Dror, J. L. Klepeis, I. T. Arkin, M. Ø Jensen, H. Xu, N. Trbovic, and R. A. Friesner, The Journal of Physical Chemistry B 112, 6155 (2008). [18] R. O. Dror, D. H. Arlow, D. W. Borhani, M. Ø Jensen, S. Piana, and D. E. Shaw, Proceedings of the National Academy of Sciences 106, 4689 (2009). [19] D. E. Shaw, P. Maragakis, K. Lindorff-Larsen, S. Piana, R. O. Dror, M. P. Eastwood, J. a. Bank, J. M. Jumper, J. K. Salmon, Y. Shan, et al, Science 330, 341 (2010). [20] V. A. Voelz, G. R. Bowman, K. Beauchamp, and V. S. Pande, Journal of the American Chemical Society 132, 1526 (2010). [21] J. W. Pitera, W. C. Swope, and F. F. Abraham, Biophysical journal 94, 4837 (2008). [22] F. Noé, C. Schütte, E. Vanden-Eijnden, L. Reich, and T. R. Weikl, Proceedings of the National Academy of Sciences 106, 19011 (2009). [23] J. Mittal and R. B. Best, Biophysical journal 99 (2010). [24] N. W. Kelley, X. Huang, S. Tam, C. Spiess, J. Frydman, and V. S. Pande, Journal of molecular biology 388, 919 (2009). [25] P. L. Freddolino and K. Schulten, Biophysical journal 97, 2338 (2009). [26] R. Day, D. Paschek, and A. E. Garcia, Proteins: Structure, Function, and Bioinformatics 78, 1889 (2010). [27] G. R. Bowman, K. A. Beauchamp, G. Boxer, and V. S. Pande, The Journal of chemical physics 131, 124101 (2009). [28] P. C. Whitford, A. Ahmed, Y. Yu, S. P. Hennelly, F. Tama, C. M. T. Spahn, J. N. Onuchic, and K. Y. Sanbonmatsu, Proceedings of the National Academy of Sciences 108, 18943 (2011). [29] T. Schlick and W. K. Olson, Science 257, 1110 (1992). [30] N. Clauvelin and W. K. Olson, Biophysical Journal 108 (2015). [31] T. Schlick, F1000 biology reports 1 (2009). [32] T. Schlick, F1000 biology reports 1 (2009). [33] G. G. Maisuradze, P. Senet, C. Czaplewski, A. Liwo, and H. A. Scheraga, The Journal of Physical Chemistry A 114, 4471 (2010). [34] A. Liwo, C. Czaplewski, S. Oldziej, and H. A. Scheraga, Current opinion in structural biology 18, 134 (2008). [35] H. Lei and Y. Duan, Current opinion in structural biology 17, 187 (2007). [36] M. L. Klein and W. Shinoda, Science 321, 798 (2008). [37] D. J. Earl and M. W. Deem, Proceedings of the National Academy of Sciences of the United States of America 101, 11531 (2004). [38] F. Sterpone, S. Melchionna, P. Tuffery, S. Pasquali, N. Mousseau, T. Cragnolini, Y. Chebaro, J. St- Pierre, M. Kalimeri, A. Barducci, et al, Chemical Society reviews 43, 4871 (2014). [39] T. Cragnolini, Y. Laurin, P. Derreumaux, and S. Pasquali, Journal of Chemical Theory and Computation (2015). [40] T. E. Ouldridge, A. A. Louis, and J. P. K. Doye, Physical Review Letters 104, 178101 (2010). [41] M. Baaden and S. J. Marrink, Current opinion in structural biology 23, 878 (2013). [42] S. A. Grigoryev, G. Arya, S. Correll, C. L. Woodcock, and T. Schlick, Proceedings of the National Academy of Sciences 106 , 13317 (2009). [43] O. Perišic;, R. Collepardo-Guevara, and T. Schlick, Journal of molecular biology 403, 777 (2010). [44] D. A. Beard and T. Schlick, Structure 9, 105 (2001). [45] A. Maritan, C. Micheletti, A. Trovato, and J. R. Banavar, Nature 406, 287 (2000). [46] T. X. Hoang, L. Marsella, A. Trovato, F. Seno, J. R. Banavar, and A. Maritan, Proceedings of the National Academy of Sciences of the United States of America 103, 6883 (2006). [47] T. X. Hoang, A. Trovato, F. Seno, J. R. Banavar, and A. Maritan, Proceedings of the National Academy of Sciences of the United States of America 101, 7960 (2004). [48] J. R. Banavar and A. Maritan, Annual Review Of Biophysics And Biomolecular Structure 36, 261 (2007). [49] I. Coluzza, PloS one 9 (2014). [50] I. Coluzza, PloS one 6 (2011). [51] A. Ahsan, J. Rudnick, and R. Bruinsma, Biophysical journal 74, 132 (1998). [52] F. Jülicher and R. Bruinsma, Biophysical journal 74, 1169 (1998). [53] L. Tubiana, A. L. Božič, C. Micheletti, and R. Podgornik, Biophysical journal 108, 194 (2015). [54] D. Marenduzzo, E. Orlandini, A. Stasiak, D. W. Sumners, L. Tubiana, and C. Micheletti, Proceedings of the National Academy of Sciences 106, 22269 (2009). [55] D. Jost, A. Zubair, and R. Everaers, Physical Review E 84, 31912 (2011). [56] P. Labhsetwar, J. A. Cole, E. Roberts, N. D. Price, and Z. A. Luthey-Schulten, Proceedings of the National Academy of Sciences 110 , 14006 (2013). [57] J. A. Cole and Z. Luthey-Schulten, Israel Journal of Chemistry 54, 1219 (2014). [58] J. Baschnagel, K. Binder, P. Doruker, A. A. Gusev, O. Hahn, K. Kremer, W. L. Mattice, F. Müller- Plathe, M. Murat, W. Paul, et al, Bridging the Gap Between Atomistic and Coarse-Grained Models of Polymers: Status and Perspectives (Springer Berlin Heidelberg, Berlin, Heidelberg, 2000), 152, p. 41. [59] T. Vettorel, G. Besold, and K. Kremer, Soft Matter 6, 2282 (2010). [60] M. Murat and K. Kremer, The Journal of Chemical Physics 108, 4340 (1998). [61] B. Capone, I. Coluzza, F. Lo Verso, C. Likos, and R. Blaak, Physical Review Letters 109, 238301 (2012). [62] B. Capone, I. Coluzza, R. Blaak, F. Lo Verso, and C. N. Likos, New Journal of Physics 15, 095002 (2013).
2 Program
Day 1 - Tuesday February 21, 2017
Session 1 • 13:00 to 14:30 - Registration
• 14:30 to 15:00 - Tamar Schlick In memoriam: Klaus Schulten
• 15:00 to 15:30 - Gerhard Hummer Modeling membrane sensing and remodeling dynamics
• 15:30 to 16:00 - Angel Garcia Free-energy landscape of a hyperstable RNA tetraloop
• 16:00 to 16:30 - Coffee Break
• 16:30 to 17:00 - Samuela Pasquali Predicting and exploring complex nucleic acids architectures through a coarse- grained model
• 17:00 to 17:30 - Angelo Rosa Chromosome organization and the physics of crumpled polymers
• 17:30 to 18:00 - Jonathan Doye Large-scale DNA simulations with oxDNA
Day 2 - Wednesday February 22, 2017
Session 2 • 9:00 to 9:30 - Doros Theodorou Multiscale molecular simulations of polymer-matrix nanocomposites
• 9:30 to 10:00 - Barbara Capone Multiscale coarse graining of polymer solutions
• 10:00 to 10:30 - Pietro Faccioli Self-consistent atomistic calculation of protein folding pathways
• 10:30 to 11:00 - Coffee Break
• 11:00 to 11:30 - Raffaello Potestio Multi-resolution modelling for biomolecular simulations
• 11:30 to 12:00 - Cristian Micheletti Pore translocation of knotted polymer chains: how friction depends on knot complexity
• 12:00 to 12:30 - Karissa Sanbonmatsu Simulating movement of the 30s head during translocation
• 12:30 to 14:30 - Lunch
Session 3 • 14:30 to 15:00 - Peter Bolhuis Multiscale simulations of patchy particle systems combining molecular dynamics, path sampling and green’s function reaction dynamics
• 15:00 to 15:30 - Dave Thirumalai Understanding RNA folding
• 15:30 to 16:00 - Marc Baaden Large-scale data exploration and analysis across biomolecular simulations
• 16:00 to 16:30 - Coffee Break
• 16:30 to 17:00 - Bert de Groot Challenges and opportunities in large scale alchemical free energy simulations.
• 17:00 to 17:30 - Wonpil Im Charmm-gui toward large-scale biomolecular and polymer simulations
Social dinner • 19:30 to 0:00 - Social Dinner
Day 3 - Thursday February 23, 2017
Session 4 • 9:00 to 9:30 - Gianluca Lattanzi Challenges in computational biophysics: from membrane proteins to biosensors
• 9:30 to 10:00 - Peter Freddolino Modeling protein-nucleic acid interactions from atomistic to cellular scales
• 10:00 to 10:30 - Modesto Orozco Advances and challenges in the simulation of DNA
• 10:30 to 11:00 - Coffee Break
• 11:00 to 11:30 - Michele Vendruscolo Structural basis for the different aggregation propensities of abeta40 and abeta42
• 11:30 to 12:00 - Chris Oostenbrink Reversible guest-host interactions from extensive simulations
• 12:00 to 12:30 - Ron Elber Electric fields across heterogeneous membranes
• 12:30 to 14:30 - Lunch
Session 5 • 14:30 to 15:00 - Ivan Coluzza Artificial chaperonins
• 15:00 to 15:15 - Ewa Anna Oprzeska-Zingrebe Interactions between a short DNA oligonucleotide and urea in the light of kirkwood-buff theory: a molecular dynamics simulation study
• 15:15 to 15:30 - Martin Goethe Prediction of protein configurational entropy (popcoen)
• 15:30 to 16:00 - Lennart Nilsson Codon recognition on the ribosome - free energy and qm/m calculations
• 16:00 to 16:30 - Coffee Break
• 16:30 to 17:00 - Helmut Grubmüller Atomistic simulation of single molecule experiments: molecular machines and a dynasome perspective
• 17:00 to 17:30 - Jeremy C. Smith Proteins: forever aging
Day 4 - Friday February 24, 2017
Session 6 • 9:00 to 9:30 - Simone Melchionna Macromolecules and hydrodynamics: a simulation approach
• 9:30 to 10:00 - Yassmine Chebaro Role of intrinsically disordered regions in the nuclear receptors architecture
• 10:00 to 10:30 - Sarah Harris Multiscale modelling of biomolecules: from atomistic molecular dynamics to the continuum limit with fluctuating finite element analysis
• 10:30 to 11:00 - Coffee Break
• 11:00 to 11:30 - Amir Lohrasebi The influence of a 2450 mhz electric field on the microtubule mechanical properties: a multi scale modeling approach
• 11:30 to 12:00 - Othmar Steinhauser Protein in reverse micelles - the dielectric approach
• 12:00 to 12:30 - Stefan Boresch Playing the devil's advocate: some challenges with respect to large-scale biomolecular simulations
• 12:30 to 13:00 - Summary & Conclusions
3 Abstracts
In Memoriam: Klaus Schulten Tamar Schlick[1] New York University
TBA
Modeling Membrane Sensing and Remodeling Dynamics Gerhard Hummer[1], Max Planck Institute of Biophysics, Frankfurt am Main, Germany
Living cells need to exert tight control over their lipid membranes, to maintain their internal structure, to guard their outside boundary, to establish potential and concentration gradients as their energy source, or to transmit signals between their compartments and to the outside. As a consequence, elaborate machineries have evolved that allow cells to sense and regulate both shapes and physical characteristics of their lipid membranes. Because of the complexity, size, and dynamic nature of these machineries, molecular modeling faces enormous challenges. To address these challenges, we have developed and used a range of coarse-graining strategies, combined with atomistic simulations and ensemble refinement methods. Coarse-grained molecular dynamics simulations shed light on how lipid saturation levels are sensed. To tackle large-scale membrane remodeling processes in ESCRT-mediated trafficking and autophagy, we integrate diverse simulation and coarse-graining procedures, from modeling of dynamic protein supercomplexes to near-continuum descriptions of the membrane dynamics.
Free-energy landscape of a hyperstable RNA tetraloop Angel Garcia[1] Los Alamos National Laboratory
We report the characterization of the energy landscape and thefolding/unfolding thermodynamics of a hyperstable RNA tetraloopobtained through high-performance molecular dynamics simulationsat microsecond timescales. Sampling of the configurationallandscape is conducted using temperature replica exchange moleculardynamics over three isochores at high, ambient, and negativepressures to determine the thermodynamic stability and the freeenergylandscape of the tetraloop. The simulations reveal reversiblefolding/unfolding transitions of the tetraloop into the canonicalA-RNA conformation and the presence of two alternative configurations,including a left-handed Z-RNA conformation and a compactpurine Triplet. Increasing hydrostatic pressure shows a stabilizingeffect on the A-RNA conformation and a destabilization of the lefthandedZ-RNA. We also explore the effect of urea on the energy landscape of this RNA hairpin. We find that, in addition to destabilizing the folded state, urea smoothens the RNA free energy landscape by destabilizing specific configurations, and forming favorable interactions with RNA bases.The concentration-dependence of the free energy (m- value) is observed to be in agreement with the results of other RNA hairpins. Additionally, analysis of the hydrogen-bonding and stacking interactions within RNA primarily show temperature-dependence, while interactions between RNA and urea primarily show concentration-dependence. Our results provide a comprehensive description ofthe folded free- energy landscape of a hyperstable RNA tetraloopand highlight the significant advances of all- atom molecular dynamicsin describing the unbiased folding of a simple RNA secondarystructure motif.
Predicting and Exploring Complex Nucleic Acids Architectures through a Coarse-Grained Model Samuela Pasquali[1], T. Cragnolini [1], D. Wales [1], L. Mazzanti [2], P. Derreumaux [2], P. Stadlbauer [3], J. Sponer [3], University Paris Descartes [1] University of Cambridge, UK, [2] Laboraitoire de Biochimie Theorique, CNRS UPR 9080, [3] Central European Institute of Technology, Czech Republic
Simulation of large biomolecules and biomolecular complexes is still an open challenge especially when interested large scale phenomena such as the folding of proteins and RNA. Atomistic simulations are limited to small systems and to short time scales even when adopting the most advanced computer technologies and sampling strategies.In order to access the long time scales and large conformational changes occurring in folding and assembly of nucleic acids, we have developed the coarse-grained model HiRE-RNA, that has proven effective to study folding, aggregation and assembly of RNA and DNA. I will briefly present the model explaining the key features of its force-fields and present an application to the study of G- quadruplexes.Guanine quadruplex (GQ) are a non-canonical DNA or RNA structure that can be formed by nucleotide sequences rich in guanine. In cells, putative GQ forming sequences are found in gene regulatory regions or at chromosome ends in the so-called telomeres that regulate the life cycle of the cell. Through a set of fixed-temperature CG simulations we generated physically plausible partially unfolded configurations for human telomeric GQs. In the simulations we were able to observe formation of partially folded intermediates and one event of interconversion between alternative configurations. Selected unfolded conformations were subject to standard all-atom simulations for refinement and further exploration of the energy basins. Through state-of-the art path sampling techniques and the CG model we have characterized the energy landscape for the experimentally resolved existing GQs structures, obtaining connections between all the known topologies. Results reveal a multi-funnel system, with a variety of intermediate configurations and misfolded states. This organization is identified with the intrinsically multi-functional nature of the system, suggesting a new paradigm for the study and classification of such molecules.
[1] Pasquali S., Derreumaux P., 'HiRE-RNA: A High Resolution Coarse-Grained Energy Model for RNA', JPhysChemB 2010 114:11957-66. [2] Cragnolini T, Laurin Y, Derreumaux P, Pasquali S. 'Coarse-Grained HiRE-RNA Model for ab Initio RNA Folding beyond Simple Molecules, Including Noncanonical and Multiple Base Pairings', J Chem Theory Comput. 2015, 11:3510-22. 11. P. Stadlbauer, L. Mazzanti, T. Cragnolini, D. J. Wales, P. Derreumaux, S. Pasquali, J. Sponer, 'Coarse-Grained Simulations Complemented by Atomistic Molecular Dynamics Provide New Insights into Folding of Human Telomeric G-Quadruplexes', J. Chem. Theory Comput., in press. DOI: 10.1021/acs.jctc.6b00667 [4] T. Cragnolini, D. Chakraborty, J. Sponer, P. Derreumaux, S. Pasquali and D. J. Wales 'Multifunctional Energy Landscape for a DNA G-Quadruplex: An Evolved Molecular Switch', sumbitted manuscript
Chromosome organization and the Physics of crumpled polymers Angelo Rosa[1], Ralf Everaers [1], Manon Valet [1], Ana-Maria Florescu [2] Scuola Internazionale Superiore di Studi Avanzati (SISSA), Via Bonomia 265, 34136 Trieste, Italy
Chromosome structure and dynamics make the objects of considerable experimental investigation. In the latest years, this approach has been paralleled by computer simulations of polymer models [1] which provide a remarkable accurate description of chromosome behavior under various conditions and over wide ranges of length- and time-scales. In this talk, I will discuss the analogy between chromosome conformations and the Physics of crumpled polymers in entangled solutions [2-5]. In particular, I will show how one can exploit this analogy (1) to explain the experimentally observed behavior of chromosomes in cell nuclei during interphase [2,6,7] and (2) to design an efficient multi-scale computational scheme for building model conformations of large chromosomes with different degrees of resolution [3,8].
[1] A. Rosa, C. Zimmer – Int. Rev. Cell Mol. Biol. 307, 275 (2014). [2] A. Rosa, R. Everaers – Plos Comput. Biol. 4, e1000153 (2008). [3] A. Rosa, R. Everaers – Phys. Rev. Lett. 112, 118302 (2014). [4] A. Rosa, R. Everaers – J. Phys. A: Math. Theor. 49, 345001 (2016). [5] A. Rosa, R. Everaers - J. Chem. Pays., accepted (2016). [6] A. Rosa, N. B. Becker, R. Everaers – Biophys. J. 98, 2410 (2010). [7] M. Valet, A. Rosa – J. Chem. Phys. 141, 245101 (2014). [8] A.-M. Florescu, A. Rosa – Plos Comput. Biol. 12, e1004987 (2016).
Large-scale DNA simulations with oxDNA Jonathan Doye[1] University of Oxford, United Kingdom
The oxDNA model is a coarse-grained model at the nucleotide level that has been designed to reproduce the structural, mechanical and thermodynamic properties of single- and double- stranded DNA [1,2,3]. Coupled with a GPU dynamics code it is able to simulate systems with tens of thousands of nucleotides.We have applied the model widely to the field of DNA nanotechnology [4]. We can routinely characterize the structural properties of full-scale DNA origamis and other nanostructures. We have also simulated the mechanisms of force-induced unravelling of DNA origami. Although the size scales accessible when simulating self-assembly are more limited, we have directly simulated the assembly of a mini-origami, and used rare- event methods to guide the design of a single-stranded DNA pyramid [5], to study key processes in the self-assembly of DNA star tiles into polyhedra and the complete path space for the assembly of a DNA tetrahedron. We have also extensively applied the oxDNA model to study biophysical properties of DNA, such as the response of duplex DNA to tensile, twist and bending stress. In particular, we were able to provide new insights into the coupling of denaturation and writhing for long, negatively supercoiled DNA [6].
[1] T.E. Ouldridge, A.A. Louis and J.P.K. Doye, J. Chem. Phys, 134, 085101 (2011) [2] P. Sulc, F. Romano, T.E. Ouldridge, L. Rovigatti, J.P.K. Doye and A.A. Louis, J. Chem. Phys, 137, 135101 (2012) [3] B.E.K. Snodin, F. Randisi, M. Mosayebi, P. Sulc, J.S. Schreck, F. Romano, T.E. Ouldridge, R. Tsukanov, E. Nir, A.A. Louis and J.P.K. Doye, J. Chem. Phys. 142, 234901 (2015) [4] J.P.K. Doye, T.E. Ouldridge, A.A. Louis, F. Romano, P. Sulc, C. Matek, B.E.K. Snodin, L. Rovigatti, J.S. Schreck, R.M. Harrison, and W.P. Smith, Phys. Chem. Chem. Phys. 15, 20395-20414 (2013) [5] V. Kočar, J.S. Schreck, S. Čeru, H. Gradišar, N. Bašić, T. Pisanski, J.P.K. Doye, and R. Jerala, Nat. Commun. 7, 10803 (2016) [6] C. Matek, T.E. Ouldridge, J.P.K. Doye and A.A. Louis, Scientific Reports 5, 7655 (2015)
Multiscale Molecular Simulations of Polymer-Matrix Nanocomposites Doros Theodorou[1] School of Chemical Engineering, National Technical University of Athens
In polymer-matrix nanocomposites the quantitative relationships between composition and size of polymer chains and nanoparticles, processing conditions, degree of dispersion of the nanoparticles, dynamics of the matrix chains, and macroscopic properties are still elusive. Molecular simulation holds great promise as a means for understanding and predicting these relationships, but faces serious challenges associated with the broad spectra of length and time scales governing nanocomposite properties. We will discuss a multiscale simulation strategy for materials consisting of nanoparticles of roughly spherical shape within amorphous polymer matrices. This strategy encompasses atomistic molecular dynamics (MD), coarse-grained connectivity-altering Monte Carlo (MC), and Field Theory-inspired Monte Carlo (FTiMC) simulations. Each level of representation invokes parameters that can be extracted from the previous (more detailed) levels, such that all predictions are ultimately based on an atomistic force field. Atomistic MD is useful for elucidating the details of molecular packing and in quantifying how segmental dynamics is affected by the presence of nanoparticles. Coarse- grained MC is of strategic importance in achieving equilibration at all length scales. By developing coarse-grained effective potentials from detailed atomistic ones via the Iterative Boltzmann Inversion method, vigorous MC sampling of the coarse-grained models with connectivity-altering algorithms, and reverse-mapping back to the atomistic level, one can generate well-equilibrated atomistic configurations to study structure and dynamics. We have applied this strategy to quantify the effects of incorporating fullerenes and silica nanoparticles on the segmental motion, glass transition, and elastic constants of long-chain atactic polystyrene (PS) [1,2]. In the FTiMC approach, polymer chains are represented as freely jointed sequences of statistical segments, and nanoparticles as single spherical entities. Polymer non- bonded interactions are taken into account via a functional of local density, while nanoparticle- polymer and nanoparticle-nanoparticle interactions are represented via integrated atomistic (Hamaker) potentials. This approach can elucidate changes in the conformation and spatial extent of polymer chains resulting from the presence of the nanoparticles. We have applied FTiMC to systems consisting of silica nanoparticles dispersed in monodisperse atactic PS. The nanoparticles carry monodisperse surface-grafted PS chains of prescribed molar mass and grafting density. Predicted scattering curves from the grafted polymer corona were validated against Small Angle Neutron Scattering measurements [3]. The rates of adsorption and desorption of polymer chains on solid surfaces are expected to play a significant role in shaping the viscoelastic properties of nanoparticle-filled melts and elastomers. As the time scales of these phenomena are very long, simulating them as infrequent events within a coarse-grained description of the dynamics would be very helpful. We will outline a new approach that combines self-consistent field theory with Kramers theory to compute rate constants for adsorption and desorption at a melt/solid interface of a subchain whose ends are constrained in space, e.g. through crosslinks or entanglements. Results from this approach have been validated via detailed MD simulations [4].
[1] G.G. Vogiatzis and D.N. Theodorou, Macromolecules 47 387 (2014). [2] I.G. Mathioudakis, G.G. Vogiatzis, C. Tzoumanekas and D.N. Theodorou, Soft Matter 12 7585 (2016). [3] G.G. Vogiatzis and D.N. Theodorou, Macromolecules 46 4670 (2013). [4] D.N. Theodorou, G.G. Vogiatzis, G. Kritikos, Macromolecules 47 6964 (2014).
Multiscale Coarse Graining of Polymer Solutions Barbara Capone[1], Ivan Coluzza, Christos Likos University of Vienna University of Vienna
Assessing properties of polymeric solutions in the semi-dilute regime, requires large scale simulations and powerful and reliable multiscale simulation methods. In this talk a full principles approach multiscale theoretical/computational framework is presented, and applied to polymeric systems with different architecture [1,2,3,4 ], and chemical composition [4,5,6]
[1]J. Chem. Phys., 127, 17 (2007) [2] Soft Matter, 10, 9601 (2014) [3]Soft Matter, 7, 5255 (2011) [4] Phys. Rev. Lett., 109, 238301 (2012) [5] Soft Matter, 6, 6075 (2010) [6]Nanoscale, 8, 3288 (2016)
Self-Consistent Atomistic Calculation of Protein Folding Pathways Pietro Faccioli[1], S. Orioli Universita' degli Studi di Trento
The Bias Functional (BF) approach [1,2] is a recently proposed reaction path sampling technique which can be used to study extremely complex conformational reactions, using state- of-the-art atomistic force fields. This technique is based on identifying the most realistic reaction pathways within an ensemble of uncorrelated trial reactive trajectories, generated by a specific form of biased dynamics, called ratchet-and-pawl MD (rMD). Indeed, a rigorous variational condition is applied to identify the trial rMD pathways which have the largest probability to occur in the absence of any bias. The advantage of the BF scheme is that it allows to attack extremely complicated problems, such as the folding of proteins consisting of hundreds of amino acids and folding time as long as several tens of minutes, using realistic force fields in both explicit and implicit solvent [1-5]. Whenever benchmarked against Anton results or against experiment, BF was found to give consistent results [1,2,3,5]. The main shortcoming of the BF method is of course that it relies on an ad-hoc choice of biasing reaction coordinate (entering the definition of rMD), which defines the model path space of the variational approximation and introduces a source of uncontrolled systematic error. In this talk I will announce a major step forward in the development of the BF approach showing how such a model dependence can be effectively and rigorously removed by determining the ideal biasing coordinate in a self- consistent way, i.e. from the molecule’s own dynamics, through an iterative procedure. We shall discuss the preliminary applications of this method to protein folding. We could rigorously show that, such a self-consistent algorithm provides a "nearly" exact sampling of the reaction paths (in some sense to be discussed in the talk). At the same it provides a self-consistenly determined reaction coordinate. References: [1] S. a Beccara, L. Fant, P. Faccioli, Variational scheme to compute protein reaction pathways using atomistic force fields with explicit solvent. Phys Rev Lett. 114, 098103 (2015).[2] S. a Beccara, T. Skrbic, R. Covino, P. Faccioli, Dominant folding pathways of a WW domain. Proc Natl Acad Sci USA 109, 2330–2335 (2012).[3] G. Cazzolli, F. Wang, S. a Beccara, A. Gershenson, P. Faccioli, P. L. Wintrode, Serpin latency transition at atomic resolution. Proc Natl Acad Sci USA. 111, 15414-15419 (2014). [4] S. a Beccara, T. Skrbic, R. Covino, C. Micheletti and P. Faccioli, Folding pathways of a knotted protein with a realistic atomistic force field. PLoS Comp. Biol. 9 (3), e1003002 (2013).[5] F. Wang, S. Cazzolli, P. Wintrode and P. Faccioli,Folding Mechanism of Proteins IM7 and IM9: Insight from All-Atom Simulations in Implicit and Explicit Solvent, J. Phys. Chem. B 120, 9297 (2016).
Multi-resolution modelling for biomolecular simulations Raffaello Potestio[1] Max Planck Institute for Polymer Research
One of the most challenging aspects in the computational study of complex biomolecules is their multi-scale nature, that is, the fact that many relevant processes take place on a broad range of length and time scales. It is often the case that the interplay between the latter prevents the usage of coarse-grained models, due to the lack of relevant chemical detail. On the other hand, the large system sizes make it difficult if not impossible to employ a highly detailed yet computationally expensive atomistic representations. Dual resolution simulation methods aim at overcoming these limitations making use of different models with different resolution in a concurrent fashion, thereby employing a detailed description where strictly necessary and an effective, coarse-grained model elsewhere. In this talk I will provide an overview of the basic techniques enabling this approach, and present their application to a selection of cases of relevance in the field of biochemistry and protein science.
Pore translocation of knotted polymer chains: how friction depends on knot complexity Cristian Micheletti[1], Antonio Suma [1], Angelo Rosa [1] [1] SISSA, Trieste, Italy
Knots can affect the capability of polymers to translocate through narrow pores in complex and counter- intuitive ways that are still relatively unexplored. We report here on a systematic theoretical and computational investigation of the driven translocation of flexible chains accommodating a large repertoire of knots trapped at the pore entrance. These include composite knots, which are the most common form of spontaneous entanglement in long polymers. Two unexpected results emerge from this study. First, the high force translocation compliance does not decrease systematically with knot complexity. Second, the response of composite knots is so dependent on the order of their factor knots, that their hindrance can even be lower than some of their prime components. We show that the resulting rich and seemingly disparate phenomenology can be captured in a seamless framework based on the mechanism by which the tractive force is propagated along and past the knots. The quantitative scheme can be viably used for predictive purposes and, hence, ought to be useful in applicative contexts, too. If time allows I will also cover recent results about the self-assembly of structures with addressable topological complexity.
A. Suma, A. Rosa and C. Micheletti, ACS Macro Letters 4 1420 (2015) G. Polles, E. Orlandini and C. Micheletti, ACS Macro Letters 5 931 (2016) G. Polles, D. Marenduzzo, E. Orlandini and C. Micheletti, Nature Communications, 6 , art. no 6423 (2015)
Simulating movement of the 30S head during translocation Karissa Sanbonmatsu[1], Wataru Nishima, Kara Fischer Los Alamos National Laboratory
Intrasubunit head movement has been identified as an essential motion required for translocation of mRNA through the ribosome. Recent single molecule FRET data has suggested that exaggerated motions of the head, beyond what are observed in structural studies, are required for the rate-limiting step of translocation. We use molecular simulations of the 70S ribosome translocation complexes to explore these exaggerated motions and identify regions of the ribosome that place constraints on the maximal displacements of the head.
Multiscale Simulations of Patchy Particle Systems Combining Molecular Dynamics, Path Sampling and Green’s Function Reaction Dynamics Peter Bolhuis[1] University of Amsterdam
Important reaction-diffusion processes, such as biochemical networks in living cells, or self- assembling soft matter, span many orders in length and time scales. In these systems, the reactants’ spatial dynamics at mesoscopic length and time scales of microns and seconds is coupled to the reactions between the molecules at microscopic length and time scales of nanometers and milliseconds. This wide range of length and time scales makes these systems notoriously difficult to simulate.While mean-field rate equations cannot describe such processes, the mesoscopic Green’s Function Reaction Dynamics (GFRD) method enables efficient simulation at the particle level provided the microscopic dynamics can be integrated out. Yet, many processes exhibit non-trivial microscopic dynamics that can qualitatively change the macroscopic behavior, calling for an atomistic, microscopic description.The recently developed multiscale Molecular Dynamics Green’s Function Reaction Dynamics (MD-GFRD) approach combines GFRD for simulating the system at the mesocopic scale where particles are far apart, with microscopic Molecular (or Brownian) Dynamics, for simulating the system at the microscopic scale where reactants are in close proximity [1]. The association and dissociation of particles are treated with rare event path sampling techniques. I will illustrate the efficiency of this method for patchy particle systems. Replacing the microscopic regime with a Markov State Model avoids the microscopic regime completely. The MSM is then pre-computed using advanced path-sampling techniques such as multistate transition interface sampling. I illustrate this approach on patchy particle systems that show multiple modes of binding.MD–GFRD is generic, and can be used to efficiently simulate reaction-diffusion systems at the particle level, including the orientational dynamics, opening up the possibility for large-scale simulations of e.g. protein signaling networks.
[1] A. Vijaykumar, P.G. Bolhuis and P.R. ten Wolde, J. Chem. Phys. 43, 21: 214102 (2015)
Understanding RNA folding Dave Thirumalai[1], Natasha Denesyuk, Naoto Hori University of Texas at Austin
Ribozymes, which carry out phosphoryl transfer reactions, often require Mg$^{2+}$ ions for catalytic activity. Catalytic Mg$^{2+}$ ions must bind at specific locations in the active site of the ribozyme. Correct folding of the active site and ribozyme tertiary structure is itself regulated by metal ions in a manner which is not fully understood. Here, we employ molecular simulations to show that individual structural elements of the group I ribozyme from the bacterium {it Azoarcus} form spontaneously in the unfolded ribozyme even at very low Mg$^{2+}$ concentrations, and are transiently stabilized by coordination of Mg$^{2+}$ ions to specific nucleotides. However, frustration due to competition for scarce Mg$^{2+}$ and topological constraints arising from chain connectivity prevent complete folding of the ribozyme.Much higher Mg$^{2+}$ concentration is required for complete folding of the ribozyme and stabilization of the active site. When Mg$^{2+}$ is replaced by Ca$^{2+}$ the ribozyme folds but the active site remains unstable. Our results suggest that group I ribozymes utilize the same interactions with specific metal ligands for both structural stability and chemical activity. The general mechanism of Mg$^{2+}$-induced folding of the group I ribozyme is likely employed in the folding of other types of non-coding RNA.
Large-scale Data Exploration and Analysis across Biomolecular Simulations Marc Baaden[1] Laboratoire de Biochimie Théorique, CNRS, UPR9080, Univ Paris Diderot, Sorbonne Paris Cité, Paris, France
Computational biology greatly benefits from approaches such as molecular dynamics simulations to study complex molecular assemblies.In this context, interactive visualization, manipulation and analysis aids hypothesis generation and exploration of large datasets. Integration of experimental data (SAXS, CryoEM, ..) in this modeling process is challenging. I will illustrate these issues through all-atom molecular dynamics simulations of membrane proteins [1] and coarse-grained HiRE-RNA simulations [2]. We make use of the UnityMol framework [3] based on the Unity3D game engine. A particular focus lies on interactive exploration and manipulation using tools such as haptic devices or head-mounted displays.
[1] M. Dreher, J. Prevoteau-Jonquet, M. Trellet, M. Piuzzi, M. Baaden*, B. Raffin*, N. Férey, S. Robert, S. Limet: "FD169: Exaviz: A Flexible Framework to Analyse, Steer and Interact with Molecular Dynamics Simulations", Faraday Discuss., 2014, 169, 119-142. doi: 10.1039/C3FD00142C. [2] F. Sterpone, S. Melchionna, P. Tuffery, S. Pasquali, N. Mousseau, T. Cragnolini, Y. Chebaro, J.-F. Saint-Pierre, M. Kalimeri, A. Barducci, Y. Laurin, A. Tek, M. Baaden, P. Hoang Nguyen, P. Derreumaux: "The OPEP coarse-grained protein model: from single molecules, amyloid formation, role of macromolecular crowding and hydrodynamics to RNA/DNA complexes", Chem. Soc. Rev., 43, 2014, 4871-4893. doi:10.1039/C4CS00048J. [3] Z. Lv, A. Tek, F. Da Silva, C. Empereur-mot, M. Chavent and M. Baaden: "Game on, Science - how video game technology may help biologists tackle visualization challenges" (2013), PLoS ONE 8(3):e57990. doi:10.1371/journal.pone.0057990 (http://unitymol.sourceforge.net)
Challenges and opportunities in large scale alchemical free energy simulations. Bert de Groot[1] Max Planck Institute for biophysical chemistry
Alchemical free energy calculations yield an unprecedented accuracy ofstabilities and affinities. Historically hampered by sampling problemsand cumbersome simulation setup, recent hardware developments and thecreation of automated simulation topologies for alchemicaltransformations yield the possibility to routinely screen vast numbersof transitions. Development of the pmx framework allowed us to carryout hundreds of protein mutations and monitor the effect on proteinstability in a set of modern molecular mechanics forcefields. Comparison to experiment highlights the thermodynamic accuracyof current force fields and shows the benefit of running simulationsin multiple force fields. Similarly, this study also reveals differentsources of remaining error, both on the simulation as well as on theexperimental side.
CHARMM-GUI Toward Large-Scale Biomolecular and Polymer Simulations Wonpil Im[1] Lehigh University
CHARMM-GUI, http://www.charmm-gui.org, is a web-based graphical user interface that prepares complex biomolecular systems for molecular simulations. CHARMM-GUI creates input files for a number of programs including CHARMM, NAMD, GROMACS, AMBER, GENESIS, LAMMPS, Desmond, OpenMM, and CHARMM/OpenMM. Since its original development in 2006, CHARMM-GUI has been widely adopted for various purposes and now contains a number of different modules designed to set up a broad range of simulations: (1) PDB Reader & Manipulator, Glycan Reader, and Ligand Reader & Modeler for reading and modifying molecules; (2) Quick MD Simulator, Membrane Builder, Nanodisc Builder, HMMM Builder, Monolayer Builder, Micelle Builder, and Hex Phase Builder for building all-atom simulation systems in various environments; (3) PACE CG Builder and Martini Maker for building coarse- grained simulation systems; (4) DEER Facilitator and MDFF/xMDFF Utilizer for experimentally guided simulations; (5) Implicit Solvent Modeler, PBEQ-Solver, and GCMC/BD Ion Simulator for implicit solvent related calculations; (6) Ligand Binder for ligand solvation and binding free energy simulations; and (7) Drude Prepper for preparation of simulations with the CHARMM Drude polarizable force field. Recently, new modules have been integrated into CHARMM-GUI, such as Glycolipid Modeler for generation of various glycolipid structures, and LPS Modeler for generation of lipopolysaccharide structures from various Gram-negative bacteria. These new features together with existing modules are expected to facilitate advanced molecular modeling and simulation thereby leading to an improved understanding of the structure and dynamics of complex biomolecular systems. In this talk, I will share these capabilities and discuss potential future directions in the CHARMM-GUI development project.
Challenges in computational biophysics: from membrane proteins to biosensors Gianluca Lattanzi[1] Department of Physics, University of Trento
The continuous development of theoretical models and computational methods has contributed to an unprecedented arsenal of different and/or complementary approaches aimed at the investigation of nature. However, only the comparison between model predictions and experimental data can ultimately assess the quality of the chosen model of reference. In this respect, biological materials constitute a challenging yet paradigmatic benchmark for the application of theoretical physics in an integrative approach that takes into account also the particularities of the involved disciplines.In this talk, I will present case studies in which theoretical models and computational methods allowed us to interpret experimental data and elucidate observed puzzling phenomena. In particular, I will focus on our recent studies on membrane proteins (AQP4 and ClC-1) and organic polymers employed in the fabrication of biosensors.In each presented case, the questions posed by the experimental investigations dictated the choice of the computational model that, in turn, lead to a progress in the understanding of the observed behavior. This could not have been achieved without a close collaboration and a continuous exchange of information between scientists with most diverse backgrounds. I will present also the unsolved challenges posed by the investigated systems, and discuss the methods we intend to apply to tackle them.
[1] G. F. Mangiatordi, D. Alberga, D. Trisciuzzi, G. Lattanzi and O. Nicolotti, Human Aquaporin-4 and Molecular Modeling: Historical Perspective and View to the Future, Int. J. Mol. Sci. 17:E1119. [2] E. Macchia, D. Alberga, K. Manoli, G. F. Mangiatordi, M. Magliulo, G. Palazzo, F. Giordano, G. Lattanzi and L. Torsi, Organic biolectronics probing conformational changes in surface confined proteins, Sci. Rep. 6:28085. [3] G. F. Mangiatordi, D. Alberga, C. D. Altomare, A. Carotti, M. Catto, S. Cellamare, D. Gadaleta, G. Lattanzi, F. Leonetti, L. Pisani, A. Stefanachi, D. Trisciuzzi and O. Nicolotti, Mind the gap! A journey towards computational toxicology, Mol. Informatics 35, 294. 2015 [4] D. Alberga, G. F. Mangiatordi, F. Labat, I. Ciofini, O. Nicolotti, G. Lattanzi and C. Adamo, Theoretical Investigation of Hole Transporter Materials for Energy Devices, J. Phys. Chem. C 119, 23890-23898. [5] D. Alberga, G. F. Mangiatordi, A. Motta, O. Nicolotti and G. Lattanzi, Effects of Different Self Assembled Monolayers on Thin-Film Morphology: a Combined DFT/MD Simulation Protocol, Langmuir 31, 10693- 10701. [6] D. Alberga, A. Perrier, I. Ciofini, G. F. Mangiatordi, G. Lattanzi and C. Adamo, Morphological and charge transport properties of Amorphous and Crystalline P3HT and PBTTT: Insights From Theory, Physical Chemistry Chemical Physics 17, 18742-18750. [7] P. Imbrici, L. Maggi, G. F. Mangiatordi, M. M. Dinardo, C. Altamura, R. Brugnoni, D. Alberga, G. Lauria Pinter, G. Ricci, G. Siciliano, R. Micheli, G. Annicchiarico, G. Lattanzi, O. Nicolotti, L. Morandi, P. Bernasconi, J.-F. Desaphy, R. Mantegazza and D. Conte Camerino, ClC-1 mutations in myotonia congenita patients: insights into molecular gating mechanisms and genotype–phenotype correlation, Journal of Physiology 593, 4181-4199. [8] G. F. Mangiatordi, D. Alberga, L. Siragusa, L. Goracci, G. Lattanzi and O. Nicolotti, Challenging AQP4 druggability for NMO-IgG antibody binding using molecular dynamics and molecular interaction fields, Biochimica et Biophysica Acta (BBA) – Biomembranes 1848, 1462-1471. [9] D. Y. Mulla, E. Tuccori, M. Magliulo, G. Lattanzi, G. Palazzo, K. Persaud and L. Torsi, Capacitance modulated transistor detects odorant binding protein chiral interactions, Nature Communications 6:6010. 2014 [10] F. Pisani, M. G. Mola, L. Simone, S. Rosito, D. Alberga, G. F. Mangiatordi, G. Lattanzi, O. Nicolotti, A. Frigeri, M. Svelto and G. P. Nicchia, Identification of a Point Mutation Impairing the Binding between Aquaporin-4 and the Neuromyelitis Optica Autoantibodies, The Journal of Biological Chemistry 289, 30578-30589. [11] D. Alberga, O. Nicolotti, G. Lattanzi, G. P. Nicchia, A. Frigeri, F. Pisani, V. Benfenati, G. F. Mangiatordi, A new gating site in human aquaporin-4: Insights from molecular dynamics simulations. Biochimica et Biophysica Acta (BBA) – Biomembranes 1838, 3052-3060. [12] D. Alberga, G. F. Mangiatordi, L. Torsi and G. Lattanzi. Effects of Annealing and Residual Solvents on Amorphous P3HT and PBTTT Films. Journal of Physical Chemistry. C, Nanomaterials and Interfaces 118, 8641-8655.
Modeling Protein-Nucleic Acid Interactions from Atomistic to Cellular Scales Peter Freddolino[1], Morteza Khabiri, Grace Kroner University of Michigan, USA University of Michigan, USA
The regulatory networks of all cells hinge on a complex web of protein-nucleic acid interactions, governing genome conformation and replication, transcription, post-transcriptional regulation, and translation. A wealth of recent high-throughput experimental approaches are providing us with massive databases highlighting the locations of protein-DNA and protein-RNA interactions in living cells under a wide variety of physiological conditions. Translating these data sets into a comprehensive understanding of the underlying logic of cellular regulatory networks, however, will require a concerted combination of experimental and modeling efforts to develop appropriate computational tools for analyzing and extending the results of experimental studies. We describe an integrated hierarchy of computational and experimental approaches aimed toward building genome-wide data sets on protein-nucleic acid interactions into a multiscale model for gene regulation. At the highest levels of detail, we show how all-atom molecular dynamics simulations provide crucial insight into interpreting the changing occupancy patterns of DNA-binding proteins under changing physiological conditions. At the same time, we have identified deficiencies in the representation of protein-DNA interactions in modern molecular dynamics force fields that hinder the applicability of these simulations to determining the DNA binding affinity landscapes of transcription factors. At lower resolutions, we are able to use thermodynamically driven mesoscale models to merge high-throughput data sets on protein occupancy and chromosome conformation to begin understanding the interplay between nucleoid protein binding and regulation of gene expression in bacterial genomes. We focus on recently discovered regions of high protein occupancy in bacterial chromosomes referred to as EPODs (extended protein occupancy domains), and show the mechanisms through which the protein occupancy of EPODs shapes the chromosome itself and controls transcriptional initiation and gene regulation.
Advances and challenges in the simulation of DNA Modesto Orozco[1] IRB Barcelona
DNA is a one pf the most remarkable examples of a multi-scale multi-physics problem. Individual interactions occur in the sub-nanometer scale (the nucleobase), while global effects involve the entire chromatin fiber, which for humans measures 2 meters for each cell. Interactions affecting the DNA happens in the femto to pico-second time scale, but impact the biology of DNA in the hours to years time scale. DNA represents then a major challenge for simulation techniques that needs to tackle this multi-scale problem by using a variety of approaches. I will summarize in my talk our efforts to develop a continuum of methodologies able to capture DNA properties from the small (electron) to the large (chromosome) scales
Structural basis for the different aggregation propensities of Abeta40 and Abeta42 Michele Vendruscolo[1] University of Cambridge
The amyloid β peptide (Aβ) is the major component of the amyloid plaques characteristically found in certain brain tissues of Alzheimer’s disease patients. It is well known that the 42- residue form of Aβ (Aβ42) has a higher tendency to aggregate that the 40-residue form (Aβ40). Although the sequence determinants of this behaviour have been established on the basis of the physico-chemical properties of the amino acids comprising the two sequences, it is still unclear whether in the monomeric forms of the two peptides there are additional conformational properties that contribute to their different aggregation propensities. To answer to this question we have determined highly accurate structural ensembles of the monomeric forms of Aβ40 and Aβ42 by using NMR chemical shifts and residual dipolar couplings as structural restraints in replica-averaged molecular dynamics simulations. Our analysis of these structural ensembles reveals that the most populated state of Aβ40 is more soluble than that of Aβ42. Furthermore also the higher free energy states of Aβ40 tend to be more soluble than the corresponding ones of Aβ42. These results thus identify the structural basis for the higher propensity to aggregate of Aβ42 with respect to Aβ40 by revealing that Aβ42 tends to populate compact conformations in which the more soluble regions of its sequence are not fully exposed to the solvent.
Reversible guest-host interactions from extensive simulations Chris Oostenbrink[1] University of Natural Resources and Life Sciences Vienna
The calculation of protein-ligand binding free energy energies is an important goal in the field of computational chemistry. Applying path-sampling methods for this purpose involves calculating the associated potential of mean force (PMF) and gives insight into the binding free energy along the binding process. Without a priori knowledge about the binding path, sampling reversible binding can be difficult to achieve. Sampling over all windows may be enhanced by a Hamiltonian replica exchange approach, but this requires special care concerning the reaction coordinate. For small molecules we introduce the distancefield (DF) as a reversible reaction coordinate. DF is a grid-based method in which the shortest distance between the binding site and a ligand is determined avoiding routes that pass through the protein. However, to extend the approach to the quantification of protein-protein interactions, we find that a construction of hidden restraints leads to most efficient calculations, albeit still requiring extensive simulations to ensure convergence.
Electric Fields Across Heterogeneous Membranes Ron Elber[1], Lauren Webb, Alfredo Cardenas, Rebika Shreshta, Cari M Anderson University of Texas at Austin University of Texas at Austin
Biological membranes are heterogeneous structures with complex electrostatic profiles arising from lipids, sterols, membrane proteins, and water molecules. We investigated the effect of cholesterol and its derivative 6-ketocholestanol (6-kc) on membrane electrostatics by directly measuring the dipole electric field (F) within lipid bilayers containing cholesterol or 6-kc at concentrations of 0-40 mol% through the vibrational Stark effect (VSE). We found that adding low concentrations of cholesterol, up to ~10 mol%, increases F, while adding more cholesterol up to 40 mol% lowers F. In contrast, we measured a monotonic increase in F as 6-kc concentration increased. We propose that this membrane electric field is affected by multiple factors: the polarity of the sterol molecules, the reorientation of the phospholipid dipole due to sterol, and the impact of the sterol on hydrogen bonding with surface water. We used molecular dynamics simulations to examine the distribution of phospholipids, sterol and helix in bilayers containing these sterols. At low concentrations, we observed clustering of sterols near the vibrational probe whereas at high concentration, we observed spatial correlation between the positions of the sterol molecules. This work demonstrates how a one-atom difference in a sterol changes the physicochemical and electric field properties of the bilayer.
[1] Rebika Shrestha, Cari M. Anderson, Alfredo E. Cardenas, Ron Elber, Journal of Physical Chemistry B, accepted.
Artificial Chaperonins Ivan Coluzza[1], Clarion Tung, Angelo Cacciuto University of Vienna Columbia University, NY
Incorrect folding of proteins in living cells may lead to malfunctioning of the cell machinery. To prevent such cellular disasters from happening, all cells contain molecular chaperones that assist nonnative proteins in folding into the correct native structure. One of the most studied chaperone complexes is the GroEL-GroES complex. The GroEL part has a "double-barrel" structure, which consists of two cylindrical chambers joined at the bottom in a symmetrical fashion. The hydrophobic rim of one of the GroEL chambers captures nonnative proteins. The GroES part acts as a lid that temporarily closes the filled chamber during the folding process. Several capture-folding-release cycles are required before the nonnative protein reaches its native state. Here we report molecular simulations that suggest that translocation of the nonnative protein through the equatorial plane of the complex boosts the efficiency of the chaperonin action. If the target protein is correctly folded after translocation, it is released. However, if it is still nonnative, it is likely to remain trapped in the second chamber, which then closes to start a reverse translocation process. This shuttling back and forth continues until the protein is correctly folded. Our model provides a natural explanation for the prevalence of double-barreled chaperonins. Moreover, we argue that internal folding is both more efficient and safer than a scenario where partially refolded proteins escape from the complex before being recaptured. Based on this results we propose a design for a device to help single protein refold and cluster break down.
Interactions Between a Short DNA Oligonucleotide and Urea in the Light of Kirkwood-Buff Theory: a Molecular Dynamics Simulation Study Ewa Anna Oprzeska-Zingrebe[1], Jens Smiatek [1] [1] Institute for Computational Physics, University of Stuttgart, Stuttgart, Germany
In nature, a wide range of biological processes, such as transcription termination or intermolecular binding, is dependent on the formation of specific DNA secondary and tertiary structures [1,2]. These structures can be both stabilized or destabilized by the small biological cosolutes, coexisting with the nucleic acids in the cellular environment. In our study, we investigate a simple 7-nucleotide DNA hairpin with the sequence d(GCGAAGC) [3] in the presence of varying concentrations of urea. The interaction between DNA and urea in unbiased molecular dynamics simulations has been analysed according to Kirkwood-Buff theory [4,5]. We implemented the local/bulk partitioning model [6], complemented by the analysis of preferential hydration and preferential interaction coefficients, to get insight into the distribution of the cosolute in the vicinity of the DNA oligonucleotide. The free energy landscape of unfolding has been approached via Metadynamics [7,8] upon the addition of a bias potential. This study allows us to get a more comprehensive understanding of the stability of the DNA structures in the presence of urea.
[1] Z. Zhang, N. I. Dmitrieva, R. L. Levine and M. B. Burg, Proc. Natl. Acad. Sci. U.S.A. 101 9491 (2004) [2] D. Japrung, M. Henricus, Q. Li, G. Maglia and H. Bayley, Biophys. J. 98 1856 (2010) [3] P. Padrta, R. Stefl, L. Králík, L. Zídek and V. Sklenár, J. Biomol. NMR 24 1 (2002) [4] J. G. Kirkwood and F. P. Buff, J. Chem. Phys. 19 774 (1951) [5] V. Pierce, M. Kang, M. Aburi, S. Weerasinghe and P. E. Smith, Cell Biochem. Biophys. 50 1 (2008) [6] D. J. Felitsky and M. T. Record, Biochemistry 43 9276 (2004) [7] A. Laio and M. Parrinello, Proc. Natl. Acad. Sci. U. S. A. 99 12562 (2002) [8] A. Laio and F. L. Gervasio, Rep. Prog. Phys. 71 126601 (2008)
Prediction of Protein Configurational Entropy (Popcoen) Martin Goethe[1], Jan Gleixner, Ignacio Fita, J. Miguel Rubi University of Barcelona, Spain
Minimizing a suitable free energy expression G is arguably the most common approach in (ab initio) protein structure prediction where the achieved accuracy depends crucially on the quality of G. Nevertheless, one mayor contribution of the free energy, configurational entropy, is basically always neglected in G because its determination requires sampling of many configurations which is computationally too expensive for most applications. This praxis, however, can lead to incorrect results as configurational entropy has major impact on the native state selection of proteins [1]. Here, we suggest a knowledge-based approach for incorporating configurational entropy [2] which is extremely fast as it does not involve any kind of sampling. Instead, entropy is predicted with an artificial neural network which was trained on simulation data [3] of ~1000 representative proteins. The method is freely available as a server application (called Popcoen) allowing to easily incorporate configurational entropy into existing protein software. This can yield a significant performance increase as we show exemplarily for a prominent protein software tool (FoldX [4]).
[1] M. Goethe, I. Fita, and J.M. Rubi, J. Chem. Theory Comput. 11, 351 (2015). [2] M. Goethe, J. Gleixner, I. Fita, and J.M. Rubi, in preparation. [3] http://mmb.irbbarcelona.org/MoDEL/, T. Meyer et al., Structure 18, 1399 (2010). [4] http://foldxsuite.crg.eu/, J. Schymkowitz et al., Nucleic. Acids. Res., W382 (2005).
Codon Recognition on the Ribosome - Free Energy and QM/M Calculations Lennart Nilsson[1], Alessandra Villa, You Xu, Yossa Dwi Hartono, Mika Ito Karolinska Institutet, Stockholm, Sweden
Classical free energy calculations have been combined with QM and QM/MM calculations of the affinitity of codon-anticodon interactions in the ribosomal decoding site, where several crystal structures [1,2] indicate that an anticodon base in the wobble position, assumes the normally high-energy enol form when it is involved in a non-cognate base pair. Our previous classical free energy calculations[3], which do not account for the cost of the keto->enol conversion, showed that the extra hydrogen bond provided by the enol form would indeed stabilize the non- cognate base pairs. We have now extended these results with QM/MM calculations in the ribosomal context.
[1] A. Weixlbaumer, F.V. Murphy, A. Dziergowska, A. Malkiewicz, F.A.P Vendeix, P.F. Agris, V. Ramakrishnan, Nat Struct Mol Biol 14 498 (2007). [2] N. Demeshkina, L. Jenner, E. Westhof, M. Yusupov, G. Yusupova, Nature 484 256(2012) [3] O. Allnér, L. Nilsson, L. RNA 17 2177 (2011 )
Atomistic Simulation of Single Molecule Experiments: Molecular Machines and a Dynasome Perspective Helmut Grubmüller[1] Max Planck Institute for Biophysical Chemistry, Göttingen
Ribosomes are highly complex biological nanomachines which operate at many length and time scales. We combined single molecule, x-ray crystallographic, and cryo-EM data with atomistic simulations to elucidate how tRNA translocation and the action of antibiotics work at the molecular level. We will show that tRNA translocation between A, P, and E sites is rate limiting, and identified dominant interactions. We also show that the so-called L1 stalk actively drives tRNA translocation, and that 'polygamic' interactions dominate the intersubunit interface, thus explaining the detailed interaction free energy balance required to maintain both controlled affinity and fast translation We will further suggest a new combined mechanism for translational stalling due to erythromycin bound in the exit tunnel. We will, finally, take a more global view on the 'universe' of protein dynamics motion patterns and demonstrate that a systematic coverage of this 'dynasome' allows one to predict protein function.
Proteins: Forever Aging Jeremy C. Smith[1] University of Tennessee/Oak Ridge National Laboratory, Oak Ridge, TN, USA
The time dependence of protein structural fluctuations is highly complex, manifesting subdiffusive, non-exponential behaviour with effective relaxation times existing over many decades in time. Using MD simulations, we have shown that, on timescales from 10-12 to 10-5 s, motions in single proteins are self-similar, non-equilibrium and exhibit ageing [1]. The characteristic relaxation time for a distance fluctuation, such as inter-domain motion, is observation-time-dependent, increasing in a simple, power-law fashion, arising from the fractal nature of the topology and geometry of the energy landscape explored. Diffusion over the energy landscape follows a non-ergodic continuous time random walk. Comparison with single- molecule experiments suggests that the non-equilibrium self-similar dynamical behaviour persists up to timescales approaching the in vivo lifespan of individual protein molecules.
1. Hu et al. Nature Physics 12, 171–174 (2016).
Macromolecules and hydrodynamics: a simulation approach Simone Melchionna[1], Mara Chiricotto [2], Fabio Sterpone[2], Philippe Derreumaux [2] Istituto Sistemi Complessi, Consiglio Nazionale delle Ricerche. ISC - CNR, Italy Laboratoire de Biochimie Théorique, IBPC - CNRS, France
The transport of macromolecules by hydrodynamic forces is relevant in biophysics and in biotechnological applications. A well known example is hydrodynamics regulating protein diffusion in crowding environments, that is, in conditions similar to the cell interior. Here diffusion is greatly reduced if compared to dilute conditions and the slowdown takes place at both long and short timescales, millisecond and nanosecond, respectively. Another typical examples is the aggregation process of peptides at the onset of Alzheimer disease, a phenomenon that is greatly accelerated by solvent-mediated interactions [1]. This class of problems involves a huge number of particles but one is mainly interested in the dynamics of a relatively small set of degrees of freedom, while detailed information on the remaining ones is unneeded. Our approach is to use a multi scale method by using an atom-based approach for the macromolecules in suspension and a kinetic-based model for the solvent [2]. I will discuss the fundamentals of the methodology and how local fluctuations and thermodynamic forces can be similarly incorporated [3]. In particular the limits of validity and the possible extensions of the simulation framework will be presented.
[1] Sterpone, F.; Melchionna, S.; Tuffery, P.; Pasquali, S.; Mousseau, N.; Cragnolini, T.; Chebaro, Y.; St- Pierre, J.; Kalimeri, M.; Barducci, A. et al. Chem. Soc. Rev. 2014, 43, 4871- 4893. [2] F. Sterpone, P. Derreumaux, S. Melchionna, J Chem Theory Comput. 2015, 11, 1843- 1853. [3] M. Bernaschi, S. Melchionna, S. Succi, M. Fyta, E. Kaxiras, J.K. Sircar, Comput. Phys. Commun. 2009, 180, 1495.
Role of intrinsically disordered regions in the nuclear receptors architecture Yassmine Chebaro[1] CNRS, Institut de Génétique et de Biologie Moléculaire et Cellulaire, Strasbourg
The retinoic acid receptor (RAR) belongs to the nuclear receptor superfamily of transcription factors and regulates genes involved in a plethora of cellular phenomena ranging from embryonic development and organogenesis to homeostasis of most adult tissues. Forming heterodimers with Retinoid X Receptors (RXRs), RARs regulate the expression of target genes via binding specific retinoic acid response elements. These receptors display the common organization of nuclear receptors, with mainly a DNA binding domain and a ligand binding domain linked by an intrinsically disordered region. Low resolution structural data has revealed a specific architecture of this heterodimer bound to their respective DNA response elements but due to their unstructured nature little is know on how the disordered hinges participate in preserving the RAR/RXR/DNA global architecture. Molecular dynamics simulations have been performed on the heterodimers and preliminary analyses suggest a crucial role of these regions in maintaining the integrity of the structure, for example through contacts between them and with the DNA.
Multiscale Modelling of Biomolecules: From atomistic Molecular Dynamics to the continuum limit with Fluctuating Finite Element Analysis Sarah Harris[1], Ben Hanson, Albert Solernou, Daniel Read, Oliver Harlen University of Leeds University of Leeds
Biophysical techniques that provide structural information at the mesoscale, such as cryo- electron microscopy and 3D tomography, are now sufficiently mature that they merit their own online repository called the EMDataBank (EMDB). We have developed a continuum mechanics description of proteins which uses this new experimental data as input to the simulations, and which we are developing into a software tool for use by the biomolecular science community. The model is a Finite Element algorithm which we have generalised to include the thermal fluctuations that drive protein conformational changes, and which is therefore known as Fluctuating Finite Element Analysis (FFEA) [1]. We will explain the physical principles underlying FFEA and provide a practical overview of how a typical FFEA simulation is set up and executed. We will then demonstrate how FFEA can be used to model flexible biomolecular complexes from EM and other structural data using our simulations of the molecular motors and protein self-assembly as illustrative examples. We then speculate how FFEA might be integrated with atomistic models to provide a multi-scale description of biomolecular structure and dynamics.
1. Oliver R., Read D. J., Harlen O. G. & Harris S. A. “A Stochastic finite element model for the dynamics of globular macromolecules”, (2013) J. Comp. Phys. 239, 147-165.
The influence of a 2450 MHz electric field on the microtubule mechanical properties: a multi scale modeling approach Amir Lohrasebi[1] Department of Physics, University of Isfahan, Isfahan, Iran.
Microtubule (MT) rigidity and response to 2450 MHz electric field were investigated, via multi scale modeling approach. For this purpose, six systems were designed and simulated to consider all types of feasible interactions between α and β monomers in MT, by using all atom molecular dynamics method. Subsequently, coarse grain modeling was used to design different lengths of MT. Investigation of effects of external 2450 MHz electric field on MT showed MT less rigidity in the presence of such field, which may perturb its functions. Moreover, an additional computational setup was designed to study effects of 2450 MHz field on MT response to AFM tip. It was found, more tip velocity led to MT faster transformation and less time was required to change MT elastic response to plastic one, applying constant radius. Moreover it was observed smaller tip caused to increase required time to change MT elastic response to plastic one, considering constant velocity. Furthermore, exposing MT to 2450 MHz field led to no significant changes in MT response to AFM tip, but quick change in MT elastic response to plastic one.
Protein in Reverse Micelles - The Dielectric Approach Othmar Steinhauser[1] University of Vienna
In this computational study we present molecular dynamics (MD) simulations of reverse micelles, i.e. nano-scale water pools encapsulated by sodium bis(2-ethylhexyl) sulfosuccinate (AOT) and dissolved in isooctane. Although consisting of highly polar components, such micro- emulsions exhibit surprisingly low dielectric permittivity, both static and frequency-dependent. This finding is well supported by experimental dielectric measurements. Furthermore, the computational dielectric spectra of reverse micelles with and without the polar protein ubiquitin are almost identical. A detailed component analysis of our simulated systems reveals the underlying mechanism of the observed dielectric depolarisation. While each component by itself would make a remarkable contribution to the static dielectric permittivity, mutual compensation leads to the observed marginal net result. This compensatory behavior is maintained for all but the highest frequencies. Dielectric model theory adapted to the peculiarities of reverse micelles provides an explanation: embedding a system in a cavity engulfed by a low dielectric medium automatically leads to depolarization. In this sense experiment, simulation and theory are in accordance.
Playing the Devil's Advocate: Some Challenges with Respect to Large- scale Biomolecular Simulations Stefan Boresch[1] University of Vienna
Clearly, large scale simulations with respect to time and system sizeare crucial to further our understanding of biomolecular and polymersystems. I would like to argue, however, that a performant moleculardynamics engine is not everything. Some examples/limitations I haveencountered in recent work are: (i) When computing alchemical freeenergy differences, one rarely (if at all) can use the fastest codepaths of simulation programs. [1] (ii) What about QM/MM calculations?(iii) Our own work aimed at overcoming the performance penalty ofQM/MM calculations by means of indirect cycles [2] relies onpost-processing / a posteriori calculations on a (very) largescale. This leads to challenges concerning the handling of largeamounts of data and distributed post-processing.
[1] To the best of my knowledge, GPU acceleration for alchemical FES is currently supported only in the commercial Desmond/Maestro/FEP+ package. Work of my own (JCC 2011, 32, 2449; doi: 10.1002/jcc.21829) illustrates possible work-arounds. [2] JPC Lett. 2015, 6, 4850; doi: 10.1021/acs.jpclett.5b02164
4 Posters
Decomposition of Proteins into Dynamic Units from Atomic Cross- Correlation Functions Paolo Calligari[1], Marco Gerolin[1,2], Daniel Abergel[2], Antonino Polimeno[1] [1] Dept. of Chemical Sciences - University of Padova
Amplitudes and time scales of internal motions in proteins can be obtained from spectroscopic observables. However, their interpretation is not straightforward and requires a plausible model of the internal dynamics, which can be based on molecular dynamics (MD) simulations, theoretical models or a combination of the two. In particular, MD simulations can be used to infer useful information on cooperative motions, which can be used afterward to model the dynamics behaviour of the protein.In the proposed method, the clustering of effective correlation times allows to decompose protein structures in terms of time-scale dependent networks of dynamically correlated local domains. This segmentation of the protein on the basis of motion time scales should provide an adaptive strategy for coarse-graining internal motions, depending on the problem (or technique) at hand, and could be used in the derivation of stochastic models for flexible macromolecules. In this context, it may serve as a basis for the development of a unified framework for the derivation of dynamic models that allow to extend the range of time scales accessed by MD simulations.This approach is of particular interest for the interpretation of spectroscopic data when dynamical processes at different time scales are probed with complementary experimental techniques.Preliminary results obtained for several prototypal proteins will illustrate this approach.
Calligari P. et al., J. Chem. Theory Comp., 2016, (in press) DOI: 10.1021/acs.jctc.6b00702
The role of membrane-mediated interactions in the process of immunological synapse formation Nadiv Dharan[1], Oded farago Department of Biomedical Engineering Ben–Gurion University of the Negev, Be'er Sheva, Israel
Cellular adhesion is achieved when specialized membrane proteins form specific receptor– ligand bonds with other molecules in the extracellular matrix, the cytoskeleton or other cell membranes. These adhesion bonds often aggregate and form large adhesion domains that provide strong cell anchoring and play a vital role in many biological processes such as cell migration, signal transduction and embryogenesis. One possible source of attractive forces between adhesion bonds is the membrane–mediated interactions. These originate from the increase in the membrane's fluctuation entropy and decrease in its curvature elastic energy, occurring upon bond condensation. Our research work analyzes the membrane-mediated mechanisms by using exact statistical–mechanical results, novel mean–field theories and computer simulations. We show that the membrane's curvature energy generates an attractive pair potential of mean force (PMF) with a characteristic range of 𠜉~50–100 nm, which can induce formation of semi–dilute adhesion domains with densities of about one bond per unit area 𠜉2. Remarkably, such densities resemble the densities of receptor-ligand bonds in the immunological synapse (IS). This suggests that membrane elasticity might have an important role in the formation of this specialized structure. However, conventional phase separation theories cannot account for the special concentric pattern observed in the IS, which requires an additional symmetry breaking mechanism. We show that the combination of the curvature- induced interactions with additional cytoskeletal effects (cytoskeleton remodeling, pinning, and actin retrograde flow) allows the centripetal aggregation of receptor-ligand bonds in the IS. Moreover, our results demonstrate that membrane-mediated interactions do not only allow the IS to form on biologically relevant timescales, but can also promote formation of peripheral micro-clusters in the early phases of the process, which are of great biological importance to signaling events that are vital to a proper immune response.
Structural ensemble of the protein Tau from molecular dynamics simulations Agusti Emperador[1], Modesto Orozco [1] Institute for Research in Biomedicine, The Barcelona Institute of Science and Technology, Barcelona, Spain [1] Institute for Research in Biomedicine, The Barcelona Institute of Science and Technology, Barcelona, Spain
We have generated the structural ensemble of the 441-residue long protein Tau with a long discrete molecular dynamics simulation, using a highly accurate coarse-grained protein model. We have analysed the structural ensemble of this intrinsically disordered protein, finding very good agreement with global and local characteristics of unbound, monomeric Tau observed in experiments: radius of gyration, flexibility along sequence and residual secondary structure. We could sample the complete conformational space of this large protein thanks to the low viscosity in the implicit model that we use, which produces a sampling orders of magnitude faster than that produced in a standard molecular dynamics simulation with atomistic resolution and explicit solvent.
Inverted signaling by bacterial chemotaxis receptors Fan Jin[1], Shuangyu Bi [1], Victor Sourjik [1] [1] Max Planck Institute for Terrestrial Microbiology, Marburg, Germany
Microorganisms use transmembrane sensory receptors to perceive a wide range of environmental factors, enabling them to adjust their physiology by changes in gene expression and actively locate optimal conditions using chemotaxis. Although behavioral responses of Escherichia coli and several other bacteria have been extensively studied, elucidating molecular mechanisms that underlie sensing and transduction of multiple stimuli by sensory receptors remains a challenge. Similarly unclear is how easily sensory properties of receptors are modified in the process of evolutionary adaptation to novel environments. In this work, we demonstrate surprising plasticity of bacterial chemotaxis receptors, showing that the sign of signaling by E. coli Tar receptor can be easily inverted by mutations at several specific sites along receptor sequence. This inversion could be reproduced by molecular dynamics simulations, shedding light on the mechanism of the transmembrane signaling by E. coli chemoreceptors. We further use receptors with inverted signaling to map molecular determinants that enable one receptor to sense multiple environmental factors, including metal ions, aromatic compounds, osmotic pressure, and salt ions.
Molecular mechanism of canonical activation of p38α MAP kinase Antonija Kuzmanic[1], Ludovico Sutto [2], Giorgio Saladino [2], Francesco L. Gervasio [2], Angel R. Nebreda [1][3], Modesto Orozco [1][4][5] [1] Institute for Research in Biomedicine (IRB Barcelona), The Barcelona Institute of Science and Technology, Barcelona, Spain [1] Institute for Research in Biomedicine (IRB Barcelona), The Barcelona Institute of Science and Technology, Barcelona, Spain [2] University College London, London, United Kingdom [3] Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain [4] Joint BSC-IRB Program in Computational Biology, Barcelona, Spain [5] University of Barcelona, Barcelona, Spain p38α is a Ser/Thr protein kinase involved in a variety of cellular processes and pathological conditions, which makes it a promising pharmacological target. Although the activity of the enzyme is highly regulated, its molecular mechanism of activation remains largely unexplained, even after decades of research. By using state-of-the-art molecular dynamics simulations, we decipher the key elements of the complex molecular mechanism refined by evolution to allow for a fine tuning of p38α kinase activity. Our study describes for the first time the molecular effects of different regulators of the enzymatic activity, and provides an integrative picture of the activation mechanism that explains the seemingly contradictory X-ray and NMR data.
CHARACTERIZATION OF COARSE-GRAINED HELIX-COIL TRANSITION KINETICS USING MARKOV STATE MODELS Joseph Rudzinski[1], Kurt Kremer[1], Tristan Bereau[1] [1] Max Planck Institute for Polymer Research
Simple, physics-based coarse-grained (CG) models have provided tremendous insight into the essential features of the protein folding process. Recent advancements in CG methodologies allow increased chemical detail and accuracy, while retaining the sampling efficiency to address problems intractable for atomically-detailed models. This beneficial speed-up, attained through a combination of reduced molecular friction and softer interaction potentials, comes at the cost of obscuring the connection to the true dynamical properties of the underlying system. Although it is possible to rescue the dynamics via a generalized Langevin formalism, this approach offers a daunting computational and conceptual challenge for complex biological molecules that give rise to hierarchical dynamics, i.e., kinetic processes coupled over various timescales. As an alternative, this work considers a Markov state modeling framework for characterizing and correcting the hierarchy of slow kinetic processes generated from CG simulations. In particular, the scheme identifies essential adjustments to a Markov state model (MSM), generated from CG simulations, in order to achieve consistent kinetics, according to a set of given external observables. We first test the method on two CG peptide models, with distinct representations and parametrization schemes, and demonstrate that the resulting information may be directly and effectively employed for model reparametrization. In both cases, the reparametrization results in an improved hierarchy of slow kinetic processes, e.g., more consistent ratios of mean- first-passage times between metastable states, while retaining the fundamental properties of the original model. In the case of a minimal structure-based model for a simple helix-coil transition, the procedure clearly identifies a general approach for parametrization: incorporating transition pathway information yields more robust simulation models. Finally, we examine more realistic helix-coil transitions and characterize the fundamental differences between atomically-detailed and native-based CG models.
[1] J.F. Rudzinski, K. Kremer and T. Bereau, J. Them. Phys. 144 051102 (2016) [2] J.F. Rudzinski and T. Bereau, Eur. Phys. J. Spec. Top. 225 1373 (2016)
A multiscale model of chromatin at bp-level Jürgen Walther[1], Pablo D. Dans [1][2], Modesto Orozco [1][2][3] [1] Institute for Research in Biomedicine (IRB Barcelona) - The Barcelona Institute of Science and Technology, Barcelona, Spain [2] Joint BSC-IRB Research Program in Computational Biology, Barcelona, Spain [3] Department of Biochemistry and Molecular Biology, University of Barcelona, Barcelona, Spain
The three dimensional organization of chromatin inside the cell nucleus is expected to strongly depend on sequence specific properties of nucleosomal and linker DNA. However, recent experiments [1] cannot capture yet the characteristics of chromatin arrangement on the resolution level of a single base-pair. To model the chromatin fiber with bp-level accuracy we first developed a coarse-grained DNA model (CG) to investigate sequence dependent DNA properties (simulation times ~10â µ times faster than conventional all atom molecular dynamics (MD) simulations). In our model, DNA is represented intrinsically at base pair level with an elastic potential representing the interactions between adjacent base pairs. Coupling terms between base pairs are extracted from atomistic MD simulations of DNA with the new parmbsc1 force field [2]. A comparison of DNA structures generated by CG and by all atom MD reveals striking similarity of important features such as distribution of helical parameters and bending properties. To extend the DNA model towards chromatin, firstly linker DNA is described as above in the CG model, secondly the nucleosome is introduced as a rigid object in between two linkers and thirdly electrostatic and steric potentials account for long-range intra-fiber interactions. This representation of chromatin makes it possible to study the characteristics of few kb-long chromatin fibers of arbitrary linker sequence and length at base pair level accuracy. Among others we find out that a long linker separating a set of nucleosome strings of equal linker lengths can play an important role in the formation of self-associating domains in the genome.
[1] Boettiger et al., Super-resolution imaging reveals distinct chromatin folding for different epigenetic states, Nature, (2016) [2] Ivani et al., Parmbsc1: a refined force field for DNA simulations, Nature Methods, (2015)
Probing drug-membrane interactions for Hypericin by large-scale molecular dynamics Pauline Walton[1], Andrea Catte [1] and Vasily S. Oganesyan [1] [1] University of East Anglia, Norwich, UK
Liposomes are a popular system for probing drug-membrane interactions and drug delivery mechanisms because they are structurally similar to biological membranes, compositionally flexible and quite easily synthesised. Hypericin is a therapeutically versatile compound isolated from Hypericum perforatum which has been found to localise preferentially in lipid membranes and tends to form pi-stack aggregates. The purpose of this work is to apply large scale state-of- the-art coarse-grained (CG) molecular dynamics (MD) simulations to a drug-loaded vesicle in order to: i) provide qualitative and quantitative insights into membrane-drug behaviour; ii) compare to previous computational and empirical studies. Inspired by Eriksson group studies, a CG vesicle (diameter 12 nm) of dipalmitoyl-phosphotidyl-choline (DPPC) was loaded with the drug Hypericin and simulated alone and then in the presence of a pure DPPC bilayer. Preliminary findings suggest that after 20 µs Hypercin aggregation at the membrane interfaces anchors and stabilises the two membrane systems. Subject to successful validation of these simulations, further tests will be undertaken to explore detailed behaviour. Since experimental evidence shows Hypericin has higher affinity for cholesterol (CHOL) over other lipids a new MD model will be built replacing the pure DPPC bilayer with a symmetric ternary bilayer containing DPPC, dioleoyl-phosphotidyl-choline, CHOL (2:2:1). This simulation will test whether the aggregation of HYP is reduced in the ternary system compared to pure DPPC.
[1] J. P. Jambeck, E. S. Eriksson, A. Laaksonen, A. P. Lyubartsev and L. A. Eriksson, J Chem Theory Comput, 10 (1), 5-13, (2014). [2] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark and H.J. Berendsen, J Comput Chem, 26 (16), 1701-18, (2005). [3] Y. Qi, H.I. Ingólfsson, X. Cheng, J. Lee, S.J. Marrink and W. Im, J Chem Theory Comput, 11, 4486- 4494, (2015). [4] W. Humphrey, A. Dalke and K. Schulten, Journal of Molecular Graphics, 14:33-38, (1996). [5] S. Jo, T. Kim, V.G. Iyer and W. Im, J Comput Chem, 29 (11), 1859-65, (2008). [6] Open Babel, version 2.3.1, http://openbabel.org (accessed May 2016). [7] Y-F. Ho, M-H. Wu, B-H. Cheng, Y-W. Chen and M-C. Shih, BBA, 1287-1295, (2009).
Ab initio Monte Carlo folding of the 92-residue alpha+beta protein TOP7 Olav Zimmermann[1], Jan Meinke [1], Sandipan Mohanty [1] [1] Jülich Supercomputing Centre, Jülich, Germany
Despite recent progress in both algorithms and available computer power there are only few reports of unbiased folding simulations for non-trivial proteins. We have recently performed High Performance Monte Carlo simulations using ProFASi and, starting from random conformations, could reproducibly simulate folding of TOP7, an ultrastable designed protein. With 92 residues this is the largest and by far most complex protein for which ab initio folding simulation has been reported to date. Due to its extreme stability and its complex folding landscape special analysis methods are required to derive its free energy landscape of folding at relevant temperatures. Our simulations are in line with the experimental results regarding stability and folding properties of TOP7 and mark significant progress in the field.
1. Mohanty, S., Meinke, J.H. & Zimmermann, O.. Proteins 81, 1446–1456, (2013). 2. Irbäck, A. & Mohanty, S.. Journal Of Computational Chemistry 27, 1548-1555, (2006). 3. Zhang, Z. & Chan, H.S.. Biophysical Journal 96, L25-L27, (2009).
5 Participant List
Organizers Capone, Barbara University of Vienna, Austria Coluzza, Ivan University of Vienna, Austria Dellago, Christoph University of Vienna, Austria Pasquali, Samuela Paris Descartes University, France Schlick, Tamar New York University, USA
Baaden, Marc - CNRS-Institute of Physico-Chemical Biology, Paris, Germany Bianchi, Emanuela - Institute for Theoretical Physics, TU Wien, Austria, Austria Bianco, Valentino - University of Vienna, Austria Bolhuis, Peter - University of Amsterdam, The Netherlands Boresch, Stefan - University of Vienna, Austria C. Smith, Jeremy - Oak Ridge National Laboratory, USA Calligari, Paolo - University of Padova, Italy Cardelli, Chiara - University of Vienna, Austria Chebaro, Yassmine - CNRS, Institut de Génétique et de Biologie Moléculaire et Cellulaire, Strasbourg, France D. Davari , Mehdi - RWTH Aachen University , Germany de Groot, Bert - Max Planck Institute for Biophysical Chemistry, Göttingen, Germany de Llano, Elisa - Austrian Institute of Technology, Austria Dharan, Nadiv - Department of Biomedical Engineering, Ben-Gurion University of the Negev, Israel Doye, Jonathan - University of Oxford, United Kingdom Elber, Ron - UT Austin, USA Emperador, Agusti - Institute for Research in Biomedicine, Barcelona, Spain Faccioli, Pietro - Physics Department of Trento University, Italy Freddolino, Peter - University of Michigan, USA Frezza, Elisa - Université Lyon 1 / UMR 5086 CNRS, IBCP, France Garcia, Angel - Rensselaer Polytechnic Institute, USA Garon, Arthur - University of Vienna, Austria Goethe, Martin - University of Barcelona, Spain Grubmüller, Helmut - Max Planck Institute for Biophysical Chemistry, Göttingen, Germany Güssregen, Stefan - Sanofi-Aventis Deutschland GmbH, Frankfurt am Main, Germany Harris, Sarah - University of Leeds, United Kingdom Hummer, Gerhard - Max Planck Institute of Biophysics, Germany Ibrahim Adam, Yagoub Ali - University of Sao Paulo, Brazil Im, Wonpil - Lehigh University, USA Ivanov, Ivaylo - Georgia State University, USA Jin, Fan - Max Planck Institute for Terrestrial Microbiology, Germany Kirmizialtin, Serdal - NewYork University Abu Dhabi , United Arab Emirates Kulczycka-Mierzejewska, Katarzyna - University of Warsaw, Poland Kuzmanic, Antonija - IRB Barcelona, Spain Lahiri, Ansuman - University of Calcutta, India Lattanzi, Gianluca - University of Trento, Italy Locatelli, Emanuele - University of Vienna, Austria Lohrasebi, Amir - University of Isfahan, Iran Melchionna, Simone - IPCF - Consiglio Nazionale delle Ricerche, Italy Micheletti, Cristian - International School for Advanced Studies (SISSA), Trieste, Italy Mohammadyarloo, Zahra - Institute for Advance Studies in Basic Science, Iran Mouhib, Halima - landoltweg 2, AC, Germany Nerattini, Francesca - Computational Physics Group - University of Vienna, Austria Nilsson, Lennart - Karolinska Institutet, Sweden Oostenbrink, Chris - University of Natural Resources and Life Sciences, Austria Oprzeska-Zingrebe, Ewa Anna - University of Stuttgart, Germany Orozco, Modesto - University of Barcelona and Institute for Research in Biomedicine, Spain Piotto, Stefano - University of Salerno, Italy Potestio, Raffaello - Max Planck Institute for Polymer Research, Germany Rosa, Angelo - International School for Advanced Studies (SISSA), Trieste, Italy Rudzinski, Joseph - Max Planck Institute for Polymer Research, Germany Sanbonmatsu, Karissa - Los Alamos Natl. Lab., USA, USA Singh, Kuldeep - MLSM College Sundernagar ,Himachal Pradesh University Shimla, India Steinhauser, Othmar - University of Vienna, Austria Theodorou, Doros - National Technical University of Athens (GR), Greece Thirumalai, Dave - University of Maryland, MD, USA Tubiana, Luca - University of Vienna, Austria Vendruscolo, Michele - University of Cambridge, United Kingdom Walther, Jürgen - IRB Barcelona, Spain Walton, Pauline - University of East Anglia, United Kingdom Wieder, Marcus - University of Vienna, Austria Zimmermann, Olav - Forschungszentrum Juelich, Germany
6 Venue
The workshop will be held at the:
Joseph Loschmidt Hörsaal Faculty of Chemistry University of Vienna. Währinger Str. 42, 1090 Wien
From the airport the cheapest way to reach the faculty is the S7 train and get down at Handelskai U-Bahn (metro) station. From there take the U6 line direction Siebenhirten and exit at Währinger Straße-Volksoper. At that point you can either walk for about 500 m or take the 40, 41 or 42 tramline direction Schottentor and exit at the second stop. The cost will be around 5 Euros.
Alternatively you can get a taxi (get it inside not outside) for about 35/40 euros. Maybe you have the car2go subscription, if so remember that it works also in Vienna and it will cost you around 10 euros to drive to the University.
Coffee breaks and poster sessions will take place at the faculty of chemistry on the floor above the conference room.
7 Hotels
The following hotels that are close to the venue of the workshop:
• Boltzmann Hotel: http://www.hotelboltzmann.at/
• Hotel Harmonie: http://www.tripadvisor.com/Hotel_Review-g190454-d227177-Reviews- BEST_WESTERN_Hotel_Harmonie-Vienna.html
• Appartement Pension 700m zum Ring: http://www.booking.com/hotel/at/700mzumring.en.html
• Hotel & Palais Strudlhof: http://www.sotour.at/wien/hotel-palais-strudlhof/
• Ibis Styles Hotel: http://ibisstyleswien.com/en/
Also good sources are websites such as: • http://www.tripadvisor.com/ • http://www.booking.com/ • http://www.venere.it/ • https://www.airbnb.com
8 Lunch Lunch options located near the faculty of physics are available at the addresses below. In addition there are several places that offer quicker options like sandwiches or take away.
• Hotel & Palais Strudlhof Pasteurgasse 1 (Austrian ~20 Euros) (~5 min walking)
• Pizza Angolo 22 Währinger Str. 22, 1090 Wien (Italian ~10 Euros) (~6 min walking)
• VinziRast Mittendrin Währinger Str. 19, 1090 Wien (Austrian ~10 Euros) (~5 min walking)
• Gasthaus Rebhuhn Berggasse 24, 1090 Wien (Austrian ~15 Euros) (~11 min walking)
9 Social Dinner
Finally the program includes a social dinner, which will take place at the Schübel-Auer Heuriger:
Kahlenberger Straße 22 1190 Wien
Directions from the workshop venue can be found on the ÖBB website. Below you can find a suggested route. Print page Close window
Connections
Overview Outward journey Station Date Time Duration Changes Products Wien Spitalgasse/Währinger Straße dep 19:02 22.02.2017 0:31 1 Walk, Tram 33, , Tram D, Walk Wien Nußdorf Beethovengang (Zahnradbahnstraße) arr 19:25 Wien Spitalgasse/Währinger Straße dep 19:09 22.02.2017 0:31 0 Walk, Tram 37, Walk Wien Hohe Warte (Wollergasse) arr 19:23 Wien Spitalgasse/Währinger Straße dep 19:10 22.02.2017 0:31 1 Walk, Tram 33, , Tram D, Walk Wien Nußdorf Beethovengang (Zahnradbahnstraße) arr 19:33
Connections due to your requested settings Travel information available only for the period from 13.12.2015 to 09.12.2017. Software/Data: HAFAS 5.41.OEBB.4.7-p6/5.41.OEBB.4.7/t-9.07 - 20.12.2016 / 14:59 / srvti 10.02.2017 / 06:54 / vfj03 © 1996-2017 ÖBB-Personenverkehr AG/HaCon Ingenieurgesellschaft mbH. No guarantee is taken for accuracy and completeness of the information given. Subject to alteration. Map basics and footpaths are adopted from routing systems. Platform/ rail information may differ due to operational requirements. Please pay attention to the local information. Otherwise apply General Terms and Conditions of Use (http://www.oebb.at/en/rechtliches/nutzungsbedingungen).