Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 1
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics National Academy of Sciences National Academy of Engineering Institute of Medicine National Research council September 2, 2010 Jodi Swidzinski Hezky, Ph.D. D. E. Shaw Research 120 West 45th Street, 39th Floor New York, NY 10036 Dear Dr. Hezky: This letter details the work and transmits the final report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics. The committee evaluated submissions in response to a Request for Proposals (RFP) for Biomolecular Simulation Time on Anton, a specialized supercomputer designed and constructed by D.E. Shaw Research that allows for dramatically accelerated molecular dynamics simulations. D.E. Shaw is making time on a 512 node Anton machine available to the non-commercial research community without cost. During this time, the Anton machine will be housed at the Pittsburgh Supercomputing Center (PSC)’s National Resource for Biomedical Supercomputing; the related support work to enable Anton to be used by the community is supported by a grant from the National Institute of General Medical Sciences to the PSC. The work of the National Research Council (NRC) committee to evaluate proposals requesting allocations of time on Anton was supported by a contract between D.E. Shaw Research and the National Academy of Sciences and was performed under the auspices of the National Research Council’s Board on Life Sciences. To undertake this task, the National Research Council convened a committee of experts to consider the proposals submitted in response to the Anton allocation RFP. The committee of 16 was chaired by Dr. Robert L. Jernigan, Director of the Baker Center for Bioinformatics and Biological Statistics and Professor of Biochemistry, Biophysics, and Molecular Biology at Iowa State University. The committee members were chosen for their expertise in molecular dynamics simulations, as well as in the subject areas represented in the 67 proposals that were considered by the committee for simulation time. The committee comprised a cross section of the biomolecular MD field, including both senior members and more junior investigators. The biographies of all of the committee members can be found in Appendix D. The goal of the RFP for Biomolecular Simulation Time on Anton is to facilitate breakthrough science in the study of biomolecular systems by providing access to a dedicated, massively parallel supercomputer that allows significantly faster simulations of biomolecular systems using periodic boundary conditions and explicit solvent models. Anton’s capabilities allow questions to be addressed on multi-microsecond simulation timescales, so the program seeks to support
OCR for page 2
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics projects addressing important and potentially high impact questions that would be most advanced by receiving time on this specialized machine. The Anton RFP and a review guidance document listed criteria against which the committee was requested to evaluate proposals, including: Scientific Merit, including potential to advance understanding of an important problem or question in the field; potential for breakthrough science resulting in new discoveries and understanding; impact that successful completion of the proposed research would have on the knowledge, methods, and current barriers in the field; and a scientifically and technologically feasible project with clear, well-developed, and appropriate goals, objectives, and approach to the proposed studies. Justification for Requested Time Allocation, including a clear and well-justified need for multi-microsecond simulation time and a clear and convincing justification that the length and number of proposed simulation runs and node-hours requested are necessary and sufficient to achieve the scientific objectives. Investigator Qualifications, including appropriate experience and training to successfully conduct the proposed studies, evidence of knowledge and prior experience with molecular simulations, and past publications. As described in the RFP, staff at PSC provided an initial review of proposal submissions for completeness and to determine whether they met technical requirements for simulation on Anton. A PSC staff member was also present as an observer during the review committee discussions in the event that additional technical specification questions arose. D.E. Shaw Research and Pittsburgh Supercomputing Center will make time on Anton available in two phases. During the first six months, 15 proposals will receive allocations of 100,000 node-hours each and during the second six months, approximately 30 proposals will receive allocations of 50,000 node-hours each. The committee was asked to identify the proposals which best met the selection criteria to receive simulation time on Anton for each of these two phases. The committee’s judgments are based on a determination of which proposals best met or exceeded the selection criteria described above and on the estimates of required simulation time provided by the applicants. In assembling the committee and conducting the review, the National Research Council sought to balance the need to include committee members with specialized expertise in the subfields of biomolecular MD (in order to evaluate the technical details of proposals), the desire of D.E. Shaw Research and the National Resource for Biomedical Supercomputing at the Pittsburgh Supercomputing Center (PSC) to allow members of the review committee to take advantage of the specialized opportunity provided by Anton and to submit proposals if they chose, and the need for a review process that was fair and credible. The following process was used to address these sometimes conflicting goals: Committee members were allowed to submit proposals1; 1 Three members of the committee submitted proposals as either Principal Investigators or affiliated faculty: Carol Post was the PI for proposal PSCA00077P, Klaus Schulten was the PI for proposal PSCA00024P, and David Beveridge was a member of the ABC Consortium that submitted proposal PSCA00033P (PI: Thomas Cheatham).
OCR for page 3
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics Prior to the committee meeting, each proposal was read by two primary reviewers and evaluated based on the criteria in the RFP. These criteria are described in Appendix C; The proposals were divided into two subgroups and any committee member who was either a Principal Investigator (PI) or affiliated faculty on a proposal was only involved in reviewing the subgroup of proposals that did not include their own2; All other committee members participated in the review and discussion of all proposals in both subgroups, except for those proposals submitted by PIs or affiliated faculty from their own institution or for which they either had a current academic or collaborative relationship, or with whom they had some past academic or personal relationship that led the committee member to voluntarily recuse him/herself and leave the room during the discussion of that specific proposal; and All committee members, except for those who were a PI or affiliated faculty on the proposals under consideration, participated in a final session on the second day to combine the results of the subgroup discussions and agree on proposals to be identified for allocations of time on Anton. Table 1 (Appendix A) lists the subgroups to which all proposals were assigned and the names of the committee members who were not involved in the review or discussion of each proposal. The NRC committee held its 2-day meeting in Washington, D.C. on July 26-27, 2010. On the first day of the meeting, members of the committee first discussed proposals from subgroup A and then proposals from subgroup B, following the general procedures described above. For each subgroup of proposals, the committee members who served as primary reviewers were asked to summarize their reviews of the proposals for which they were responsible, and these summaries were followed by discussion among the group. Committee members considered the relative strengths of the proposals and worked toward reaching consensus on which ones best met the Anton selection criteria. If a decision could not be made during this round of discussion, an additional committee member was assigned to review the proposal in detail. Committee members then returned to the discussion at a later time. On the second day, all committee members, except for those who were either a PI or affiliated faculty on the proposals, participated in a final session to combine the results of the subgroup discussions and agree on proposals to be identified for allocations of time on Anton. Although the committee was motivated by its desire to enable as many members of the community as possible to have the opportunity to receive time, it particularly sought to identify those proposals which it judged were most likely to lead to high impact or breakthrough science, even if these entailed some risk of failure. The proposals listed below addressed important biological questions and were those which the committee judged took best advantage of Anton’s specialized capabilities to address questions over multi-microsecond simulation timescales. The committee debated the relative advantages of using Anton to address kinetic versus equilibrium questions. In many cases, the proposals identified by the committee as best meeting the selection criteria dealt with kinetic questions and incorporated Anton’s ability to generate dynamic trajectories; they frequently involved informative comparisons between experimental and computational results. In other cases, the committee judged that proposals addressing important thermodynamic issues or aspects of methods and force-field validation would be of significant practical value to the community and would also be greatly advanced by receiving time on Anton. Conversely, some of the proposals which were evaluated as meeting the Anton selection criteria less well were judged by the 2 Dr. Post reviewed proposals in subgroup A; Drs. Beveridge and Schulten reviewed proposals in subgroup B.
OCR for page 4
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics committee to lack focus, to pose questions that needed additional preliminary research and development, or that might be alternatively met using other computational resources or analysis methods. The committee has concluded that the proposals listed below best meet the objectives of the RFP for Biomolecular Simulation Time on Anton. More detailed comments for all 67 proposals are included in Appendix B. Phase 1 (initial six months): The committee was asked to identify 15 proposals that best met the criteria of the RFP to receive approximately 100,000 node-hours each of simulation time on Anton, for a total time allocation of approximately 1,500,000 node-hours. The 44 proposals that initially requested approximately 60,000-100,000 node-hours of time were evaluated by the committee for Phase 1. Of these, the proposals listed below were judged by the committee as being the 15 that exceeded the Anton selection criteria and would most benefit from receiving full allocations of requested time. In numerical order by proposal submission number, these are: PSCA00007 Capturing Large-Scale Structural Transitions in Membrane Transporters at Atomic Resolution; PI: Emad Tajkhorshid, University of Illinois at Urbana-Champaign PSCA00010 Structural Determinants of Flickering in K+ Channels; PI: Benoit Roux, University of Chicago PSCA00013 The Atomistic Scale Kinesin Mechanism Elucidated on the Experimental Time Scale; PI: Wonmuk Hwang, Texas A&M University PSCA00023 Visualizing the Pathway of Integrin Headpiece Opening Induced by Ligand Binding; PI: Timothy Springer, Harvard University PSCA00024 Determining the Pathway of Nascent-Protein Insertion through the Protein-Conducting Channel and into the Membrane; PI: Klaus Schulten3, University of Illinois at Urbana-Champaign PSCA00028 Long Time Simulations of Protein Folding: a Synergistic Approach; PI: Vijay Pande, Stanford University PSCA00046 Toward gaining insights into the mechanism of substrate transport by the aspartate transporter GltPh; PI: Ivet Bahar, University of Pittsburgh PSCA00047 Long Timescale Molecular Dynamics Simulation of Protein Folding; PI: Martin Gruebele, University of Illinois at Urbana-Champaign PSCA00048 Computational Design and Evaluation of Novel Enzyme Catalysts; PI: Kendall Houk, University of California-Los Angeles 3 Dr. Schulten is a member of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics.
OCR for page 5
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics PSCA00052 Sequencing DNA using MspA; PI: Aleksei Aksimentiev, University of Illinois at Urbana-Champaign PSCA00059 Study of microsecond time scale protein dynamics crucial for phosphorylation-mediated signaling; PI: Michael Hagan, Brandeis University PSCA00065 Atomistic modeling of the resting and activated states of a voltage-gated potassium channel voltage-sensing domain; PI: Alfredo Freites, University of California-Irvine PSCA00072 Understanding Sec-facilitate protein translocation and membrane integration: New mechanistic insights from microsecond-timescale trajectories on Anton; PI: Thomas Miller, California Institute of Technology PSCA00074 Resolving the Molecular Mechanisms of Calcium Binding to Cadherins Involved in Hearing and Deafness; PI: David Corey, Harvard Medical School PSCA00077 Investigation of conformational properties of residues near 5-fold symmetry axis during uncoating of rhinovirus capsid by long time scale molecular dynamics simulation using Anton at NRBSC; PI: Carol Post4, Purdue University Phase 2 (second six months): The committee also was asked to identify approximately 30 proposals that best met the criteria of the RFP to receive approximately 50,000 node-hours each of simulation time on Anton, for a total time allocation of approximately 1,500,000 node-hours. The proposals identified by the committee for inclusion under this phase of the project fell into three categories: Proposals that initially requested approximately 50,000 or fewer node-hours of simulation time and that the committee judged would most benefit from a full allocation of requested time. Proposals that initially requested more than 50,000 node-hours of simulation time. The committee concluded that these proposals well met the criteria of the RFP and addressed questions that would benefit from receiving simulation time on Anton. However, these proposals were evaluated by the committee as meeting the goals of the RFP for Biomolecular Simulation Time on Anton slightly less well that the proposals selected for inclusion in Phase 1. The committee concluded that valuable progress could still be made in addressing the goals of these proposals with a reduced allocation of time of 50,000 node-hours each. Additional proposals that the committee concluded would also benefit from receiving simulation time on Anton, ranging from 15,000 to 50,000 node-hours, to complete the available allocation of simulation time. 4 Dr. Post is a member of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics.
OCR for page 6
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics A. Proposals initially requesting approximately 50,000 node-hours of time The committee judged that these proposals also exceeded the Anton selection criteria and should be considered for a full allocation of requested simulation time of 50,000 node-hours. In numerical order by proposal submission number, these are: PSCA00014 Dynamic coupling and binding in a GTPase-effector complex; PI: Matthias Buck, Case Western Reserve University PSCA00015 Understanding the mechanics of energy conversion in Na+-dependent co-transporters; PI: Michael Grabe, University of Pittsburgh PSCA00029 Linking Structure and Conductance of Ion Channels; PI: Andrew Pohorille, University of California-San Francisco PSCA00034 Molecular Dynamics Simulation of Signal Transduction in the Squid Rhodopsin G-Protein Coupled Receptor; PI: Douglas James Tobias, University of California-Irvine PSCA00037 Dynamics and functional motions of the GlpG intramembrane protease; PI: Ana Nicoleta Bondar, University of California-Irvine PSCA00057 Lipid-mediated assembly of membrane proteins; PI: Stephen White, University of California-Irvine PSCA00058 Application of New Tools for Characterizing Protein Dynamics to Microsecond-Scale, Explicitly-Solvated MD Simulations of Intrinsically Disordered and Natively Folding Proteins; PI: Michael Colvin, University of California-Merced PSCA00061 Exploring Lipid-Protein Interactions Using Microsecond-scale Molecular Dynamics Simulation; PI: Toby Allen, University of California-Davis PSCA00062 To knot or not to knot: slipknotting in the smallest knotted protein; PI: Jose Onuchic, University of California-San Diego PSCA00067 Development and testing of improved fixed-charge force fields for proteins; PI: David Case, Rutgers University B. Proposals initially requesting more than 50,000 node-hours of simulation time, to be considered for a modified allocation of 50,000 node-hours. The committee judged that these proposals well met the Anton selection criteria. However, because only 15 proposals could receive the maximum allocation of simulation time, the committee has included them among those proposals to be considered for a modified allocation of 50,000 node-hours. In numerical order by proposal submission number, these are: PSCA00001 Detailed Characterization of the Equilibrium Fluctuations of the Engrailed Homeodomain; PI: Christopher Langmead, Carnegie Mellon University PSCA00026 Characterization of Dynamics Control of Chemotaxis Initiation; PI: Jerome Baudry, University of Tennessee at Knoxville
OCR for page 7
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics PSCA00027 Molecular Flexibility in Drug Design Using Microsecond Molecular Dynamics; PI: James Andrew McCammon, University of California-San Diego PSCA00033 Molecular Dynamics of DNA and Protein DNA Complexes: A Proposal for Obtaining Micro-second Trajectories using Anton; PI: Thomas E. Cheatham5, University of Utah PSCA00068 Mechanistic insights into the “inside-out” signaling of integrins; PI: Marta Filizola, Mount Sinai Medical Center C. Additional proposals that would benefit from receiving time allocations on Anton, some with modified allocations, to complete the available allocation of simulation time The committee judged that these proposals met the Anton selection criteria and would benefit from receiving simulation time on Anton ranging from 15,000 to 50,000 node hours. In numerical order by proposal submission number, these are: PSCA00005 Exploration of the Human Adenovirus Protease Activation Pathway via Long Timescale Molecular Dynamics Simulations; PI: Ross Walker, University of California-San Diego [identified for 50,000 node-hours] PSCA00006 Continuous Long-Time Dynamics of RNA Molecules: Watching without Blink-ing for Microseconds through Anton’s Microscope; PI: Ioan Andricioaei, University of California-Irvine [identified for 50,000 node-hours] PSCA00009 Simulations of a Sterol Transport Protein (Osh4) that Tethers Membranes of the Endoplasmic Reticulum and Plasma Membrane; PI: Jeffery B. Klauda, University of Maryland at College Park [identified for 25,000 node-hours] PSCA00012 Using ANTON to probe the conformational space of poly-glutamine and its aggregation to understand its role in Huntington’s disease, a protein aggregation disease; PI: Bruce J. Berne, Columbia University [identified for 50,000 node-hours] PSCA00017 Understanding the Transcriptional Regulation of MerR; PI: Jerry M. Parks, Oak Ridge National Laboratory [identified for 50,000 node-hours] PSCA00020 Aβ Peptides Interactions with Lipid Bilayers: Implications for Aggregation and Neurotoxicity; PI: Jie Zheng, University of Akron [identified for 50,000 node-hours] PSCA00036 Uncovering the mechanism of Protein Tyrosine Phosphatase 1B induced insulin resistance in Type 2 Diabetes Mellitus through MD simulations of the PTP1B-IRK complex; PI: Thanh Truong, University of Utah [identified for 15,000 node-hours] PSCA00039 Micro-Second Molecular Dynamics Simulation of the Folding Pathways of Tetratricopeptide Repeat Units in the Cargo Binding Domain of Kinesin Motor Proteins; PI: Carol Parish, University of Richmond [identified for 50,000 node-hours] 5 Dr. Beveridge, a member of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics, is a member of the Consortium that submitted this proposal.
OCR for page 8
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics PSCA00040 Molecular Dynamics Simulations of Conformational Dynamics in the p38a MAP Kinase: Differential Dynamics in the Crystal and Solution Environments and the Effects of Binding a Small Molecule Inhibitor; PI: Adrian H. Elcock, University of Iowa [identified for 50,000 node-hours] PSCA00041 Growth mechanisms of amyloid fibrils; PI: Jian-Min Yuan, Drexel University [identified for 50,000 node-hours] PSCA00042 Characterizing changes in the conformation and dynamics of epidermal growth factor receptor induced by mutations associated with anti-cancer drug treatment; PI: Peter Coveney, Yale University [identified for 50,000 node-hours] PSCA00043 Structural and Functional Water Dynamics in Rhodopsin Activation from Picoseconds to Microseconds; PI: Mark R. Chance, Case Western Reserve University [identified for 50,000 node-hours] PSCA00045 All Atom Molecular Dynamics Simulation of Connexin Hemichannel Voltage Gating; PI: Thaddeus A. Bargiello, Albert Einstein School of Medicine [identified for 24,000 node-hours] PSCA00049 Assessment of Multi-Microsecond Simulations of Intrinsically Disordered Proteins Using NMR: Applications to FCP1 in the Unbound State; PI: Scott A. Showalter, Pennsylvania State University [identified for 50,000 node-hours] PSCA00063 All-Atom Molecular Dynamics Simulations of S-Adenosyl Methionine (SAM) Assisted SAM-I Riboswitch (Un)Folding Pathways: A Small Molecule with a Strong Arm; PI: Shantenu Jha, Louisiana State University [identified for 50,000 node-hours] PSCA00064 Structural mechanism of integrin activation induced by talin; PI: Cheng Zhu, Georgia Institute of Technology [identified for 50,000 node-hours] PSCA00073 The Effect of α-Synuclein on Membrane Structure; PI: Jonathan Sachs, University of Minnesota [identified for 50,000 node-hours] The time allocations for the 47 proposals judged by the committee as best meeting the Anton selection criteria total approximately 3,000,000 node-hours. In carrying out its charge, the committee identified as many promising proposals as feasible given the overall constraints on simulation time, to allow the maximum number of users possible the opportunity to take advantage of this special resource. If additional node-hours of simulation time become available during the course of the project, the committee encourages D.E. Shaw Research and Pittsburgh Supercomputing Center to allocate these hours to the research groups showing the most promising results and which would benefit most greatly from the additional time. The committee also encourages D.E. Shaw and PSC to provide a collective repository and to encourage investigators to share the data generated, since the trajectories obtained may be of use to other investigators in the community.
OCR for page 9
Report of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics The committee wishes to thank D.E. Shaw Research, Pittsburgh Supercomputing Center, and all of the 2010 Anton applicants for the opportunity to review these proposals and to identify which of the proposals best met the requirements set forth in the RFP. The committee members were vocal in their enthusiasm for the computational opportunities provided by Anton and pleased to be a part of helping determine how to allocate time on this important resource. Sincerely, Robert L. Jernigan Chair cc: Dr. Joel Stiles, Pittsburgh Supercomputing Center Dr. Warren Muir, National Research Council Dr. Frances Sharples, National Research Council Appendixes: A. Table 1: Proposals Reviewed by the Committee B. Individual Proposal Summary Evaluations C. Proposal Evaluation Criteria D. Roster and Biographical Sketches of Committee Members E. The Board on Life Sciences and the National Academies F. Acknowledgement of Report Reviewer