Biomolecule Benchmarking Meeting

Date

1/20/2021

These notes are now shared outside confluence to be accessible to a larger group: https://docs.google.com/document/d/1b9h8MWCAcp7zpE_BWshL0-zwcqZBT33bXU3rGkPCR74/edit

Participants

• Michael Shirts

• David Mobley

• Michael Gilson

• John Chodera

• Karmen Condic-Jurkic

• David Cerutti (Deactivated)

• Simon Boothroyd

Goals

Define biomolecular force field benchmarks

Invitees:

Scientific Advisory Board:

• David Case (Rutgers)

• Lillian T. Chong (University of Pittsburgh)

• Alex Mackerell (University of Maryland)

• Julia Rice (IBM Research)

• Sereina Riniker (ETH Zurich)

• Benoît Roux (University of Chicago)

• Carlos Simmerling (Stony Brook University)

• Bill Swope (IBM Research)

• Junmei Wang (University of Pittsburgh)

People with letters from our grant:

• Gabe Rocklin (Northwestern)

• Tom Cheatham (U of U)

• Adrian Roitberg

• Bernie Brooks

• Erik Lindahl

• Emad Tajkhorshid

• Samuli Ollia (U of Helsinki) (if we include lipids now)

Industry people:

• Relay

• Which others are interested?

Other interested parties:

• Kresten Lindorff-Larsen

• Some D. E. Shaw people?

Mechanisms for compiling data sets

• Start with a phone call to brainstorm data sets

  • Potentially “prime the pump” with a brainstorm of existing data sets and plans.

  • Lay out criteria for inclusion: must be an open data easily and automatically accessible to others

  • OpenForce Field is not required to use exactly this dataset - we may only use a subset (it WILL be a subset, since we will propose anything we plan to use).

  • Allow people who can’t make the phone call help brainstorm on documents.

• Iterate virtually for a set amount of time on documents describing the data set.

  • Documents include description of the data sets, and how they will be stored, distributed, and documented.

• Have another phone call to finalize data set

  • Have a specific mechanism for compiling and distributing the data.

Discussion topics

Action items

Decisions