Summary

This document describes the work and transmits the decisions of the Committee on Proposal Evaluation for Allocation of Supercomputing Time for the Study of Molecular Dynamics, Fifteenth Round.

The committee evaluated submissions received in response to a Request for Proposals (RFP) for Biomolecular Simulation Time on Anton 3, a supercomputer designed and built by D. E. Shaw Research (DESRES). Over the past 14 years, DESRES has made an Anton or Anton 2 system housed at the Pittsburgh Supercomputing Center (PSC) available to the non-commercial research community, based on the advice of previous committees of the National Academies of Sciences, Engineering, and Medicine (the National Academies). This year, Anton 2 will be replaced by a 64-node third-generation Anton machine, Anton 3. This new machine greatly enhances the capabilities of the previous generation Anton systems. As in prior rounds, the goal of the fifteenth RFP for simulation time on Anton 3 is to continue to facilitate breakthrough research in the study of biomolecular systems by providing a massively parallel system specially designed for molecular dynamics (MD) simulations. These capabilities allow for multi-µs simulation timescales. The program seeks to continue to support research that addresses important and high-impact questions demonstrating a clear need for Anton’s special capabilities.

The success of the program has led DESRES to make the Anton 3 machine housed at PSC available for approximately 330 machine days (MDs) over the period following February 2025, and DESRES asked the National Academies to again facilitate the allocation of time to the non-commercial community. The work of the committee to evaluate proposals for time allocations was supported by a contract between DESRES and the National Academy of Sciences and was performed under the auspices of the National Academies’ Board on Life Sciences.

To undertake this task, the National Academies convened a selection committee of experts to evaluate the proposals submitted in response to the RFP. The committee of 22 was chaired by Dr. Brian Dominy, an Associate Professor of Chemistry and Associate Dean at Clemson University. The committee members were selected for their expertise in MD simulations and experience in the subject areas represented in the 63 proposals that were considered. The members comprised a cross-section of the biomolecular dynamics field in academia, including both senior and junior investigators.

The Anton 3 RFP described the three criteria against which the committee was asked to evaluate proposals:

• Level of Scientific Merit, including the potential to advance understanding on an important problem or question in the field; the potential for breakthrough science resulting in new discoveries and understanding; the impact that successful completion of the proposed research would have on knowledge, methods, and current barriers in the field; and a scientifically and technologically feasible project with clear, well-developed, and appropriate goals, objectives, and approaches to the proposed studies.
• Justification for Requested Time Allocation, including 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 and Past Accomplishments, including the appropriate experience and training to successfully conduct the proposed studies, evidence of knowledge and prior experience in molecular simulations, and past publications.

Proposals from investigators who had previously received an allocation of time on Anton or Anton 2 were required to include progress reports, which the committee drew on as supplemental material in its consideration of proposals. As explained in the RFP, staff at PSC conducted an initial assessment of all proposal submissions for completeness and to determine if they were technically feasible for simulation on

Anton 3. Two members of the PSC staff were present as observers during committee discussions to address any additional questions that arose on the technical capabilities of Anton 3 or on how the computer will be made available to researchers during the period of the project.

The committee was asked to identify proposals that best met the selection criteria defined above. The maximum time allocation available per proposal was 9.1 MDs. Principal investigators (PIs) could also request a smaller time allocation. The committee was permitted, at its discretion, to consider a modified time allocation than otherwise requested. The committee was further asked to target, to the extent possible, at least 25% of allotted time to PIs who had not previously received an Anton allocation. The judgments of the committee are based on which proposals best met the selection criteria described above and on the estimates of required simulation time provided by the applicants.

Initial reviews of the proposals were provided by the 22 committee members. Each proposal was assigned a minimum of two primary reviewers who were asked to evaluate the proposal based on the RFP and the guidelines described above. Review assignments were made so that proposals were not evaluated by reviewers from the applicant’s same institution or reviewers who had close collaborative or personal relationships with an applicant. These conflicts of interest (COIs) are included in Appendix H.

The committee held one virtual meeting and one hybrid proposal review meeting on December 11^th and 16^th, 2024, respectively. For the hybrid meeting, some committee members attended in person in Washington, DC while others joined virtually. At the meeting, the two primary reviewers were asked to summarize their reviews for the committee, which was followed by discussion of the proposed research. Committee members determined to have a COI with a proposal were excused from the committee discussion on that proposal and placed in a virtual waiting room (if virtual) or left the room (if in person) until discussion of the proposal was complete. As described in detail above, committee members considered the scientific merit, justification of the requested time, and the qualifications of the PI and key personnel. The committee reviewed the slate of proposals under consideration and selected proposals it judged best met the selection criteria, and, in some cases, decided to suggest a modified allocation of time on Anton 3. Detailed comments for each of the 63 proposals are included in Appendix B.

The committee judged that the proposals listed below best met the selection criteria set forth in the RFP for Biomolecular Simulation Time on Anton 3. Of these 50 proposals, 18 proposals were selected for a modified allocation (identified below with an *). In alphabetical and numerical order by proposal submission number, the proposals judged by the committee as best meeting the selection criteria of the RFP are:

CHE100024P Kendall Houk, University of California, Los Angeles; Determining How Flexible Enzymes BruB and PfB Catalyze Pericyclic Cascades that Convert Conjugated Hydrocarbon Polyenes into Complex Polycyclic Natural Products [Returning user identified for 3 MDs]*

CHE140036P Jianing Li, Purdue University; Microsecond MD Simulation to Understand Mammalian AC1 Protein Activation toward Drug Discovery [Returning user identified for 5.7 MDs]

IBN130013P Maria Bykhovskaia, Wayne State University; Protein Machinery Regulating Synaptic Vesicle Fusion [Returning user identified for 6.08 MDs]

MCB100016P Aleksei Aksimentiev, University of Illinois at Urbana-Champaign; All-Atom Structure of Dengue Virus Nucleocapsid [Returning user identified for 8.52 MDs]

MCB100017P Emad Tajkhorshid, University of Illinois at Urbana-Champaign; Conformational Dynamics of the Voltage Sensing Domain During hERG Activation [Returning user identified for 9 MDs]

MCB110005P Douglas James Tobias, University of California, Irvine; Atomistic modeling of the pharmacological inhibition of the Hv1 proton channel and the adhesion function of aquaporin 0 [Returning user identified for 9.1 MDs]

MCB110012P Jeffery B. Klauda, University of Maryland, College Park; Isoform-Specific Gating Mechanisms in HCN Channels: Molecular Dynamics Simulations of cAMP and Voltage-Dependent Conformational Changes [Returning user identified for 9.1 MDs]

MCB110023P Matthias Buck, Case Western Reserve University; Computer Simulation of the Association of Full-length EphA2 RTK at the Membrane [Returning user identified for 5.2 MDs]

MCB110024P Marta Filizola, Icahn School of Medicine at Mount Sinai; Investigating the Molecular Basis of Ligand Efficacy in GPCR-Mediated G Protein Signaling [Returning user identified for 9.1 MDs]

MCB120079P Rommie Amaro, University of California, San Diego; Cytosolic Phospholipase A2 Long Timescale Molecular Dynamics and Interactions with the RAW Macrophage Endoplasmic Reticulum Membrane [Returning user identified for 9 MDs]

MCB130052P Themis Lazaridis, City University of New York; Mechanism of membrane remodeling by the caveolin 8S complex [Returning user identified for 5 MDs]*

MCB140052P Richard W. Pastor, National Institutes of Health; Structure of low-density lipoprotein [Returning user identified for 9.1 MDs]

MCB150024P Marcos Sotomayor, University of Chicago; In-Silico Mechanics of Experimental and AlphaFold 3 Models of the Vertebrate Inner-Ear Mechanotransduction Apparatus [Returning user identified for 7.89 MDs]

MCB160079P Sharon Loverde, City University of New York; Molecular Dynamics Simulations of Pioneer Transcription Factors Bound to the Nucleosome [Returning user identified for 4.2 MDs]

MCB160080P Michael Feig, Michigan State University; Inter- and intramolecular diffusion of intrinsically disordered peptides inside biomolecular condensates [Returning user identified for 9.1 MDs]

MCB160087P Mahmoud Moradi, University of Arkansas at Little Rock; Elucidating the Conformational Dynamics of Multidrug Resistance-Associated Protein 1 and P-Glycoprotein Using Microsecond-Level All-Atom Molecular Dynamics Simulations [Returning user identified for 8.93 MDs]

MCB160089P Igor Vorobyov, University of California, Davis; Multi-microsecond-long molecular dynamics simulations to assess genetic variation of cardiac ion channel – drug interactions for digital twins [Returning user identified for 9 MDs]

MCB180078P Xiaolin Cheng, Ohio State University; Investigation of Conformational Dynamics in Cell-Cell Adhesion Cadherin-Catenin-Vinculin Complex [Returning user identified for 8.4 MDs]

MCB180080P Ira Kurtz, University of California, Los Angeles; Roles of PIP2 and intracellular loop 5 in the interactions between the cytoplasmic and transmembrane domains during elevator transport in SLC4 proteins [Returning user identified for 9.1 MDs]

MCB180081P Liqun Zhang, University of Rhode Island; Anton simulation on GPCR receptors CXCR4 and AR interaction with ligands [Returning user identified for 4 MDs]*

MCB190044P Yun Luo, Western University; Gating and lipid regulation of Piezo2 channel [Returning user identified for 9.1 MDs]

MCB190070P Alexey Ladokhin, University of Kansas; Atomistic modeling of cancer-targeting peptide pHLIP on membrane interfaces [Returning user identified for 9.1 MDs]

MCB190073P Jodi Hadden, University of Delaware; Hepatitis B virus (HBV) capsid: Characterizing a viral shapeshifter through the computational microscope [Returning user identified for 8 MDs]*

MCB200071P Denise Okafor, Pennsylvania State University; Investigating ligand-specific mechanisms in full length nuclear receptors [Returning user identified for 8 MDs]

MCB200085P Jing Li, University of Mississippi; Activation and Desensitization Mechanisms of GABAA Receptors and the Modulatory Role of Glycosylation [Returning user identified for 4 MDs]*

MCB210009P Alemayehu Gorfe, University of Texas Health Science Center at Houston; Dynamics of lipid-anchored small GTPases [Returning user identified for 6 MDs]*

MCB210013P Alessio Accardi, Weill Cornell Medical College; Molecular mechanisms of gating of the human chloride channel CLC-1 and of the lipid scramblase Xkr4 [Returning user identified for 9.1 MDs]

MCB210016P Jessica Swanson, University of Utah; Unraveling GPAT4 Dynamics: Long-Timescale Simulations of Protein-Membrane Interactions in Lipid Metabolism [Returning user identified for 5.02 MDs]

MCB210020P Michael Regnier, University of Washington; Modeling departure from the interacting heads motif state of myosin [Returning user identified for 9.1 MDs]

MCB210021P Sandhya Kortagere, Drexel University; Conformational analysis of phosphorylated Dopamine transporter and its interactions with Dopamine D3 receptor [Returning user identified for 2 MDs]*

MCB220003P Alexander Neimark, Rutgers University; Multiscale Modeling of Coronavirus Virions in the Respiratory System [Returning user identified for 5 MDs]*

MCB220011P Juan Vanegas, Oregon State University; Membrane Modulation of GPCR Interactions with G-Protein and Beta-arrestin partners [Returning user identified for 3.6 MDs]*

MCB230000P Jie Xiao, Johns Hopkins University; Conformational Dynamics Governing Enzymatic Processivity of the Core Bacterial Septal Cell Wall Synthesis Complex [Returning user identified for 9.1 MDs]

MCB230004P Linda Columbus, University of Virginia; Determinants of lipid reorganization by bacterial membrane proteins: The role of protein sequence and conformation [Returning user identified for 4.1 MDs]

MCB240000P Kenton Swartz, National Institutes of Health; A Computational Search For PI(4,5)P2 Binding Sites on the Open State Kv2.1 Ion Channel [New user identified for 2 MDs]*

MCB240001P Carter Butts, University of California, Irvine; Effects of a Dimerization and Membrane Environment on Interaction with the SARS-CoV-2 Polyprotein by Common Variants of the Main Protease [New user identified for 3.84 MDs]

MCB240004P Alvin Yu, University of California, Irvine; Long-timescale simulations of molecular processes in viral propagation and biological energy [New user identified for 4.5 MDs]*

MCB240005P Anna Tarakanova, University of Connecticut; Investigating the effects of missense mutations in elastin with molecular dynamics [New user identified for 4.5 MDs]*

MCB240006P Mohtadin Hashemi, Auburn University; Effect of curvature on the membrane interactions of amyloid beta [New user identified for 5 MDs]*

MCB240008P Rozita Laghaei, Carnegie Mellon University; Computational Study Of A First-In-Class, Positive Allosteric Modulator That Targets Cav2.1 Voltage-Gated Calcium Channels [New user identified for 4.35 MDs]*

MCB240009P Yanxin Liu, University of California, San Fransisco; Molecular mechanism of client protein remodeling by the Hsp90 molecular chaperone [New user identified for 9.1 MDs]

MCB240010P David N. Beratan, Duke University; Electron Bifurcation Gating by Domain Conformational Changes in StnABC transhydrogenase [New user identified for 9 MDs]

MCB240011P Steve Cannon, University of California, Los Angeles; Simulated deactivation of the NaV1.4 sodium channel to define the proton leak through the voltage-sensor domain of a mutant channel causing periodic paralysis [New user identified for 9.1 MDs]

MCB240012P Itay Budin, University of California, San Diego; Uncovering mechanisms of ligand release and membrane selectivity in ORP/Osh phospholipid transporters [New user identified for 9.1 MDs]

MCB240014P Rachelle Gaudet, Harvard University; Investigating lipid-protein interactions of insect gustatory receptors [New user identified for 6 MDs]*

MCB240017P Premila Samuel Mohan Dass, University of Illinois at Urbana-Champaign; Diversity of macromolecular crowding effects across human cell [New user identified for 7 MDs]

MCB240019P Shanlong Li, University of Massachusetts Amherst; Phase Separation of RNA CUG Repeats [New user identified for 3 MDs]*

MCB240020P Andrew Harris, Rutgers University; Dynamics, Structural and Energetic Basis of Connexin Hemichannel Docking and Regulation by Ca²⁺ [New user identified for 3 MDs]*

MCB240021P Elizabeth Rhoades, University of Pennsylvania; The Regulatory Role of Tau’s N-Terminal Domains on Tubulin Binding and Polymerization of Microtubules [New user identified for 8.6 MDs]

MCB240022P Ravinder Abrol, California State University, Northridge; Structural Mechanisms of Receptor-Mediated G Protein Activation [New user identified for 6 MDs]*

The time allocations for the 50 proposals identified by the committee as best meeting the selection criteria for time allocations total approximately 333 MDs. Approximately 28.2% of MDs were allocated to 16 proposals whose PIs have not received time on Anton or Anton 2 (identified as “new users”). Approximately 71.8% of the MDs were allocated to proposals from PIs who have received allocations of time on Anton 2 in previous rounds (identified as “returning users”).

In carrying out its task, the committee identified as many promising proposals as possible and made difficult decisions on the number of allocations recommended given the constraints on the total available simulation time. The total simulation time requested by the submitted proposals was more than 488 MDs. As a result, not every proposal could be recommended for an allocation and many interesting projects received recommendations for less than the full allocation amount.

The committee would like to thank DESRES, PSC, and all the 2024 Anton 3 applicants for the opportunity to assist in identifying the proposals best meeting the selection criteria for time allocations on the new Anton 3 machine. The committee members are universally enthusiastic about the potential advances in the field that are facilitated by Anton 3 and are looking forward to seeing the important new results from the Anton 3 users.