Our research focuses on applying cutting-edge machine learning and deep learning techniques to tackle critical problems in life sciences. Specifically, we develop innovative frameworks to improve our understanding and prediction of drug interactions and adverse reactions, utilizing diverse data sources. Key areas of interest include Graph Convolutional Networks, Graph Embedding, and Multimodal Learning, which offer powerful tools for modeling complex biological relationships.
In addition, we are exploring advanced machine learning methods, including transfer learning, to identify key genes, modules, and pathways involved in cancer, aiming to provide new insights into cancer biology.