Megha Khosla

The paper develops a multitask transfer learning framework to predict virus-human protein interactions using statistical protein representations from a large sequence corpus, showing improved performance on benchmark datasets.

The paper evaluates miRNA-disease association prediction models, identifies data leakage issues, and improves performance on two datasets.

This study evaluates 11 feature selection methods across 27 biomedical datasets to determine their performance based on data complexity and provides recommendations for method selection.