Robert L. Jernigan

The paper develops a framework to discover new protein paralogs by integrating sequence, structure, and embedding-based methods with family-specific models, identifying 7 new protease and 3 new kinase paralogs.

The paper identifies many new human paralogs using a combined sequence, structure, and language model approach, validating catalytic residues and expanding protein functional understanding.

The paper evaluates variant effect predictions for ARSA gene variants, finding that a bootcamp-trained model outperformed others and that deep learning slightly improved performance over simpler methods.

ESMStabP predicts protein thermostability with improved accuracy using embeddings from ESM2 and thermophilic classifications, achieving high R2 and PCC values.

PLAST uses protein language models and discrete cosine transforms to efficiently detect homologs with high accuracy and speed.

The paper introduces a new sequence matching method, ProtSub, that incorporates structural information to identify novel functions for HIV gp120 proteins, showing improved accuracy over conventional methods.

The paper introduces a new amino acid similarity matrix that improves sequence matching by incorporating structural information, resulting in better sequence-structure congruence.