AI-Driven Drug Discovery, Immunoinformatics, Digital Health

We are committed to the field of artificial intelligence (AI) for bioinformatics computing and digital health. By advancing the development and application of cutting-edge deep learning and large model technologies, we aim to tackle the complex diseases that humans face in their pursuit of high-quality health, such as obesity, diabetes, and tumors. Utilizing AI to the multi-omics and clinical big data, we strike to elucidate the high-order links between cellular phenotypes and underlying molecular mechanism. Meanwhile, we built generative large language models to accelerate the de novo design of small molecule drugs, antibodies and vaccines.

Recent Works


cycleCDR: cycle consistency learning for predicting single-cell cellular responses, ISMB2024 (Published in Bioinformatics)

Wei Huang, Hui Liu*

PDF | CODE


Learning Cross-Domain Representations for Transferable Drug Perturbations on Single-Cell Transcriptional Responses, AAAI2025 (Accepted)

Hui Liu*, Shikai Jin

PDF | CODE


Identifying drug-resistant individual cells within tumors by semi-supervised transfer learning from bulk to single-cell transcriptome, Communications Biology, 2025

Kaishuan Huang, Hui Liu*

PDF | CODE


UnifyImmun: a unified cross-attention model for predicting antigen immunogenicity, Nature Machine Intelligence, 2025

Chenpeng Yu, Xing Fang, Shiye Tian, Hui Liu*

PDF | CODE | Web Server


tcrLM: a lightweight protein language model for predicting T cell receptor and epitope binding specificity

Chenpeng Yu, Xing Fang, Hui Liu*

PDF | CODE


Drug2TME: Cross-domain disentanglement for modeling tumor microenvironment impact on drug efficacy, JBHI, 2024

Jia Zhai, Hui Liu*

PDF | CODE