Prediction of protein substitution impact using a directional substitution matrix and homolog alignments

DeMaSk uses a simple linear model to predict the functional impact of single amino acid substitutions. It uses an interpretable, directional amino acid substitution matrix computed from deep mutational scanning datasets, as well as per-position evolutionary conservation and variant frequency computed from homologs of the query sequence. more

Daniel Munro and Mona Singh (2020). DeMaSk: a deep mutational scanning substitution matrix and its use for variant impact prediction. Bioinformatics, 36(22-23), 5322-5329.

For questions, email dmunro@health.ucsd.edu or see the Singh Lab website.

Enter a sequence here and then view and download the computed predictions.

— or —

Install the Python package to run DeMaSk locally.