DeMaSk

DeMaSk predicts the quantitative impact of amino acid substitutions on protein function using deep mutational scanning (DMS) datasets.

Key Features:

• Directional substitution matrix inference: Infers directional amino-acid substitution matrices from deep mutational scanning (DMS) datasets and sequence homologs.
• Linear model integration: Fits a linear model that integrates substitution scores with per-position evolutionary conservation metrics and variant frequency across homologs.
• Performance: Achieves top-tier accuracy in predicting the functional impact of missense mutations despite methodological simplicity.

Scientific Applications:

• Variant impact prediction: Predicts the functional consequences of missense mutations on protein function.
• Protein sequence analysis: Applicable to any protein sequence for assessing mutation effects using DMS-derived substitution matrices and conservation metrics.

Methodology:

From DMS datasets and sequence homologs, DeMaSk infers directional amino-acid substitution matrices and fits a linear model that incorporates substitution scores, per-position evolutionary conservation metrics, and variant frequency across homologs.