Jason Harris

This paper introduces SHERPA, a pan-allelic MHC-peptide prediction algorithm, which improves neoantigen discovery by integrating immunopeptidomics data and composite modeling.

The paper identifies genomic loci with systematic sequencing bias in clinical whole genomes and proposes a statistical method to flag suspect variants for quality control.

This study evaluates variant annotation consistency across tools and databases, finding significant discordance in variant descriptions, particularly for insertions and deletions.