Avinash D Sahu

The paper introduces MutQA, a cross-validated Q&A dataset linking genetic mutations to their functional consequences, and demonstrates its utility in improving LLM performance.

This study finds that higher pretreatment γδ T-cell abundance is linked to better outcomes in cancer patients receiving immune checkpoint blockade and improved survival across multiple tumor types.

DeepVul predicts gene essentiality and drug response using a multi-task transformer model trained on cancer transcriptome data, improving precision oncology by identifying treatment vulnerabilities.

MedGraphNet integrates multi-relational biological data with text knowledge to improve biomedical predictions, outperforming single-relation methods and identifying drug side effects without direct training.

The paper introduces scDist, a tool for detecting perturbed cell types in single-cell RNA-seq data, showing it outperforms existing methods in identifying transcriptomic changes in disease contexts.

LitGene integrates textual information into gene representations using a transformer-based model with contrastive learning, improving disease risk gene prediction and interpretability.

The paper identifies 71,946 genetic interactions in cancer using a novel method, revealing non-synthetic-lethal interactions that explain driver gene specificity and drug response.