Collin J. Tokheim

The paper develops a machine learning model to predict protein degradability for targeted protein degradation, identifying key features and extending predictions to the proteome.

The paper identifies two classes of transcription factors based on their regulatory range and shows that long-range TFs' activity depends on chromatin state.

OpenCRAVAT provides an open-source platform for variant and gene prioritization by integrating diverse data resources and computational methods.

CHASMplus identifies cancer-specific driver missense mutations across 32 cancer types using TCGA data, revealing over 4,000 unique mutations in 240 genes.

CRAVAT 4 provides tools for analyzing and interpreting somatic mutations in cancer studies, offering annotation, prioritization, and visualization with a web interface.

This study evaluates machine learning methods for predicting peptide-MHC Class I binding affinities, finding that MHCnuggets performs similarly to existing methods but is faster and handles variable-length peptides.