We are dedicated to advancing computational biology to improve our understanding of biomolecular systems and accelerate discovery in health and life sciences. We combine biological insight with data-driven methods, including deep learning, large language models, knowledge graphs, and optimization algorithms, to address challenging problems at the interface of computation and biology.
In drug discovery, our research spans method development for peptide and small-molecule drug discovery and property prediction, from target identification to lead discovery. Previous projects have focused on antimicrobial peptides (AMPs), anticancer peptides (ACPs), antioxidant peptides (AOPs), bacterial targets (TxSS), and cancer targets (TROP2 and HER2).
In health sciences, we work on medical imaging diagnosis, with a particular focus on infectious diseases, as well as spatial gene expression prediction for cancer research.
Explore our app portal (https://app.cbbio.online). Some program source codes can be downloaded from our GitHub page or SourceForge page.
Join us!!! PhD admission 2026/2027 (Admission period: 15 Oct 2025 to 15 May 2026) 「中文版」
We are proud to share that Jianxiu Cai successfully defended her dissertation on Thursday (Apr 23, 2026). Her work, “Computational Modeling of Peptide Bioactivities Using Sequence-based Deep Learning Architecture”, designed both large and slim deep learning models to address the challenges of peptide activity prediction, generating highly accurate models despite small datasets, high structural flexibility…
Read more…
Congratulations to Xiangyu on the publication of the new work in Engineering Applications of Artificial Intelligence! In this paper, the team proposes CrossToGene, a bidirectional cross-modality interaction framework for predicting spatial gene expression from histopathological images. The method introduces spatial encoding to capture both macro/micro tissue cues, a bidirectional cross-modal interaction module for stronger fusion…
Read more…
Congratulations to Xinpo Lou on the publication of his two recent works in Journal of Chemical Information and Modeling and Journal of Cheminformatics! The proposed MambaTransDTA model introduces a novel hybrid Mamba–Transformer architecture for drug–target binding affinity prediction. By effectively combining long-range dependency modeling with local interaction learning, the model achieves strong and consistent performance…
Read more…
Something went wrong. Please refresh the page and/or try again.
Antimicrobial Peptide Prediction Server
Discover the antimicrobial potential of your peptides, perform genome screening to identify novel AMPs.
Anticancer Peptide Prediction Server
Discover the anticancer potential of your peptides on six major types of cancers!
What’s next?
Want to be a designer of the next AI generation tools for therapeutics peptides, come and join us!
Chemical-Protein Target Prediction
Which proteins your ligand might bind with? Search with our AI tools.
Simple Chemical Toxicity Prediction
Our first regression model of the chemical toxicity prediction for acute oral toxicity.
Advanced Chemical Toxicity Prediction
Toxicity prediction against 12 endpoints, 7 nuclear receptor signals and 5 stress response indicators.
Accelerating drug discoveries through AI and molecular simulations
Research | Publications | Software | Team
Social Media
AIDD@Macao Polytechnic University
R. de Luís Gonzaga Gomes, Macao, China
Computational Biology and Bioinformatics 