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

Abstract

The 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. It does so by predicting 3D structural dynamics for every variant using the Elastic Network Contact Model (ENCoM), a sequence-sensitive coarse-grained normal mode analysis model. Dynamical Signatures represent the fluctuation at every position in the biomolecule and are used as features fed into machine learning models of the user's choice. Once trained, these models can be used to predict experimental outcomes for theoretical variants. The whole pipeline can be run with just a few lines of Python and modest computational resources. The compute-intensive steps are easily 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 maturation efficiency of human microRNA miR-125a variants from high-throughput enzymatic assays. Availability DynaSig-ML is open-source software available at https://github.com/gregorpatof/dynasigml_package Supplementary information Supplementary data are available at Bioinformatics online.

Understanding and predicting ligand efficacy in the mu-opioid receptor through quantitative dynamical analysis of complex structures

“…We utilized DynaSig-ML [22] to calculate entropic signatures (ES) (a vector of entropic mean square fluctuations representing the flexibility of the complex) of all ligand-receptor complexes to test the predictive value of a dynamics-derived efficacy predictor (see subsection 3.4). The predictor was constructed using LASSO linear regression [27] and the perfomance was evaluated using two cross validation tests: leave-one-ligand-out (LOLO) and leave-one-cluster-out (LOCO) (see subsection 3.5).…”

“…ENCoM has been employed to study microRNA maturation [20], the emergence of SARS-CoV-2 variants [21] as well as the determinants of thermophile protein stability [18], among others. The recently introduced DynaSig-ML Python package [22] facilitates the combination of ENCoM with simple machine learning models to capture quantitaty dynamics-activity relationships (QDAR) emerging from large datasets of experimental measures. In the present work, we use the DynaSig-ML methodology to study µ -opioid receptor activation as explained by ligand-induced changes in receptor dynamics.…”