Back

The DynaSig-ML Python package: automated learning of biomolecular dynamics-function relationships

Mailhot, O.; Major, F.; Najmanovich, R.

2022-07-07 biophysics
10.1101/2022.07.06.499058 bioRxiv
Show abstract

SummaryThe DynaSig-ML ("Dynamical Signatures - Machine Learning") Python package allows the efficient, user-friendly exploration of 3D dynamics-function relationships in biomolecules, using datasets of experimental measures from large numbers of sequence variants. The DynaSig-ML package is built around the Elastic Network Contact Model (ENCoM), the first and only sequence-sensitive coarse-grained NMA model, which is used to generate the input Dynamical Signatures. Starting from in silico mutated structures, the whole pipeline can be run with just a few lines of Python and modest computational resources. The compute-intensive steps can also easily be parallelized in the case of either large biomolecules or vast amounts of sequence variants. As an example application, we use the DynaSig-ML package to predict the evolutionary fitness of the bacterial enzyme VIM-2 lactamase from deep mutational scan data. Availability and implementationDynaSig-ML is open source software available at https://github.com/gregorpatof/dynasigml_package Contactrafael.najmanovich@umontreal.ca

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.