Avinash Yaganapu

Paper (preprint and later journal publication) introducing BIN-PU, a PU learning framework to predict bacterial compound–protein interactions using only known positive samples; includes experimental validation and publicly available code.

PhD dissertation proposing interpretable weakly supervised learning frameworks (Multi-SA for semantic segmentation, BIN-PU for PU learning in CPI prediction, BICAN-HT for cellular senescence) and a review of AI in cellular senescence research.

A narrative review summarizing AI applications in cellular senescence research and discussing computational advancements, challenges, and future opportunities.

Journal article presenting a Multi-SA approach that leverages intermediate feature representations and multi-layered self-attention to improve Class Activation Maps for weakly supervised semantic segmentation (published in Computer Vision and Image Understanding).

Master's thesis exploring GWAS data and machine learning models (ridge regression, SVM, random forest, MLP) to identify SNPs associated with bone loss rate.