/
Protein re-fit with smee

Protein re-fit with smee

 

Driver

 

Approver

 

Contributors

 

Other stakeholders

@Chapin Cavender

Objective

Fit a force field to protein data using smee.

Time frame

 

Key outcomes

  • a force field fit to protein data using smee

  • a force field to protein and small molecule data using smee

Key metrics

  • performance compared to protein QM datasets

  • folded protein benchmarks

  • small molecule benchmarks

Status

Not started In progress Stalled Complete

GitHub repo

Slack channel

Designated meeting

Released force field

Publication

 

 Problem Statement and Objective

Fitting torsion parameters to a combination of protein and small molecule QM data using ForceBalance has not yielded success. Smee may give different results.

Objective: re-fit a force field using protein data, and a combination of protein and small molecule data, using smee.

 Scope

Must have:

  • a force field trained to small molecule and protein data

  • a force field trained just to protein data

Must have:

  • a force field trained to small molecule and protein data

  • a force field trained just to protein data

Nice to have:

  • additional protein data, perhaps computed through Meta’s eSEN models, of larger peptide chains

Not in scope:

  •  

Project Approaches

References