Mackenzie M. Johnson

This study uses deep learning to detect entrenchment in antibody germline genes, revealing epistatic constraints from both intra-chain and genetically uncoupled partners.

This paper introduces a model that separates mutation and selection processes in antibody language models to improve prediction of mutation effects on antibody function.

This study compares nucleotide context models and protein language models for predicting antibody affinity maturation, finding nucleotide-based models perform better.

The paper develops 'thrifty' models of somatic hypermutation using convolutions on 3-mer embeddings, showing they outperform 5-mer models with fewer parameters.

This study quantifies sex-differential selection (SDS) across the human genome using UK Biobank data, revealing polygenic signals and a tradeoff between viability and fecundity.

The paper evaluates how different node addition mechanisms affect the resilience of PPI networks across species, finding that gene-expression-based attachment best preserves resilience.