Skip to content

presto #

Fast parameterisation of MM force fields using MLPs.

Modules:

  • analyse –

    Functionality for analysing the results of a presto run.

  • convert –

    Convert OpenFF ForceField <--> smee TensorForceField.

  • create_types –

    Create new tagged SMARTS parameter types for molecules of interest.

  • find_torsions –

    Functionality for finding and sampling torsions in a molecule.

  • hessian –
  • load_molecules –

    Helpers for loading molecules from parameterisation inputs.

  • loss –

    Functionality for computing the loss.

  • metadynamics –

    The code below is slightly modified from the original in OpenMM.

  • mlp –

    Functionality for creating and managing ML potentials.

  • models –

    Compiled models.

  • msm –

    Functionality for applying the modified Seminario method.

  • outputs –

    Functionality for handling the outputs of a workflow.

  • sample –

    Functionality to obtain samples to fit the force field to.

  • sampling_coordinator –

    Coordinate non-resumable, molecule-level sampling on one node.

  • settings –

    Pydantic models which control/validate the settings.

  • train –

    Apply OpenFF parameters to molecule, cluster conformers by RMSD and train.

  • utils –

    Utilities for the presto package.

  • workflow –

    Implements the overall workflow for fitting a bespoke force field.

  • writers –

    Output functions for run-fit.