Hua Ren Wu

DTSyn predicts synergistic drug combinations using a dual-transformer model, achieving high ROC AUC scores and outperforming existing methods.

BatchDTA improves drug-target affinity estimation by reducing batch effects through implicit batch alignment, enhancing model performance across multiple datasets.

X-MOL uses large-scale pre-training on 1.1 billion molecules to improve molecular analysis tasks through fine-tuning, achieving state-of-the-art results in property prediction, reaction analysis, and molecule generation.