Tools

MISTIC Feature selection and explanation for support vector machines

Python framework for building interpretable support vector machine ensembles, with reproducible cross-validation, kernel-aware feature selection, perturbation ranking, calibrated probability analysis, and local integrated-gradient explanations in one workflow.

FFT Protein Interpretable machine learning of protein function using Fourier-based encoding

Python package for developing and analyzing machine learned models of protein function and properties that leverage periodicities in physicochemical properties along protein sequences.

PyBio3D Protein structural analysis and generation of graph-based models of protein structure

Provides basic tools for manipulating PDB files, including visualization through dash.bio, and a framework for developing graph neural networks based on protein structural data.

Smolyak Python tools for surrogate modeling using the Smolyak sparse grid method.

Work in collaboration with Mike Howard (Auburn University) aimed at developing tools for multi-scale modeling based on Smolyak sparse grids.

introCSB R resources for learning computational systems biology

These resources include interactive modules that are generally applicable to learning the basics of systems biology modeling, as well as, coding examples for using R to develop and analyze various types of systems biology models.