Virtual Sites Phase I

 

Driver

@Lily Wang

Approver

@Brent Westbrook (Unlicensed) @Alexandra McIsaac

Contributors

@Lily Wang , @Simon Boothroyd , @Joshua Horton

Other stakeholders

@David Mobley @Michael Gilson @Michael Shirts @Daniel Cole

Objective

A force field with a set of general virtual sites

Time frame

Q1 2025

Key outcomes

  • A force field including:

    • Sigma hole virtual sites for Cl, Br

    • Virtual sites for pyridine

    • Re-trained BCCs

    • Re-trained vdW terms

    • Re-trained valence terms

  • (For comparison) a force field including:

    • Re-trained BCCs

    • Re-trained vdW terms

    • Re-trained valence terms

Key metrics

  • Improved agreement to QM ESPs

  • Improved performance on dimer energy benchmarks (DES370K)

  • Equivalent or improved performance on condensed-phase properties (NIST)

  • Improved performance on SFEs (FreeSolv and MNSol)

  • Equivalent or improved performance on valence benchmarks

Status

In progress

GitHub repo

Slack channel

 

Designated meeting

FF Fitting (Wednesdays 1pm PT)

Released force field

Publication

 

 Problem Statement and Objective

Atom centred charges do not accurately capture the anisotropy around certain moieties, such as the sigma hole present on halogens. The introduction of off-centre charges (virtual sites) can alleviate this alleviate this issue.

Our objective is to fit a set of virtual sites to electrostatic properties at HF/6-31G* to model more complex electrostatics. In the initial phase, we plan to tie this with our existing AM1-BCC charge model for the real atom particles.

 Scope

Must have:

  • Virtual sites for Cl, Br, and pyridine

  • Retrained vdW terms and valence terms

Must have:

  • Virtual sites for Cl, Br, and pyridine

  • Retrained vdW terms and valence terms

Nice to have:

  • Potentially further iodine training sites

Not in scope:

  • Other virtual sites (e.g. sulfur)

  • Higher levels of theory

Project Approaches

References