Gowri Nayar

The paper shows homology-based variant-effect predictors fail for CYP enzymes and proposes an improved model that still lacks clinical relevance.

The paper identifies high attention sites in protein language models that predict functional and pathogenic regions in proteins.

GATSBI improves protein embeddings by using biologically relevant data splits, showing better performance for understudied proteins in various tasks.

This paper identifies High Attention (HA) sites in protein sequences using the ESM model to improve protein family classification and function prediction.

This study evaluates LLMs' ability to infer biochemical reactions and disease associations from protein pathways, finding poor performance and introducing a benchmark dataset.

Pool PaRTI improves protein sequence representations by using PageRank on transformer attention matrices to identify critical residues, enhancing performance and interpretability across ML tasks.

The paper predicts epitope candidates for SARS-CoV-2 by analyzing epitope sequences from related coronaviruses and identifying conserved and diverse antigen regions across 62k genomes.

The paper develops a semi-supervised pipeline to annotate SARS-CoV-2 genomes, identifying molecular targets with high accuracy and scalability.

The paper analyzes how SARS-CoV-2 mutations affect RT-PCR primer specificity and finds a 2% monthly increase in mismatches and geographic-specific virus detection.

This paper analyzes antimicrobial resistance assay data to identify biases and contradictions in training data for machine learning models.