Advances in molecular simulation methods and hardware have finally made possible the simulation of biomolecules on physiological timescales. Such simulations provide an important tool to aid the molecular understanding of disease. However, fully integrating these new tools into medical research requires that simulation software becomes easier to use, easier to develop for, and more hardened, reliable and robust. Making these capabilities accessible and reliable in the hands of a broad scientific community will require transformative gains in reliability, interoperability, flexibility, and ease of use atscale. We therefore propose improvements to the open-source Gromacs molecular simulation package to help meet these challenges. These improvements will 1) enable automated tests of physical validity for both code contributions and simulation parameter choices, 2) provide a simple interface for external code to utilize Gromacs functionality, and 3) provide a high-performance interface to mix Monte Carlo and molecular dynamics simulation. This work will facilitate community extension via a robust testing framework integrated with our code review system and via an API to allow implementation of new tools and simulations using high-level calls. These efforts will leverage best practices from software engineering and be implemented in a way to put tools in the hands of a broad developer and user community.

Public Health Relevance

Biomolecular simulation is increasingly becoming an integral part of biomedical research. This research will harden the Gromacs molecular simulation package to create robust tests for reliability and physical validity. It will encourage broader use contributions to the Gromacs code base and ease custom modifications via a stable programming interface. Finally, it will enable these tasks to take advantage of a range of modern computing paradigms, both 'traditional' parallel computing and MapReduce and related technologies.

Agency
National Institute of Health (NIH)
Institute
National Institute of General Medical Sciences (NIGMS)
Type
Research Project (R01)
Project #
5R01GM115790-02
Application #
9266801
Study Section
Biodata Management and Analysis Study Section (BDMA)
Program Officer
Ravichandran, Veerasamy
Project Start
2016-05-01
Project End
2020-04-30
Budget Start
2017-05-01
Budget End
2018-04-30
Support Year
2
Fiscal Year
2017
Total Cost
Indirect Cost
Name
University of Virginia
Department
Physiology
Type
Schools of Medicine
DUNS #
065391526
City
Charlottesville
State
VA
Country
United States
Zip Code
22904
Kasson, Peter M; Jha, Shantenu (2018) Adaptive ensemble simulations of biomolecules. Curr Opin Struct Biol 52:87-94
Irrgang, M Eric; Hays, Jennifer M; Kasson, Peter M (2018) gmxapi: a high-level interface for advanced control and extension of molecular dynamics simulations. Bioinformatics 34:3945-3947