Hi, I’m Bo Li (李波), a Ph.D. student in Computer Science at the University of Macau, advised by Prof. Bob Zhang and co-advised by Prof. Qianqian Song (Purdue University). Since June 2026 I have been a visiting student at the School of Computing, National University of Singapore, hosted by Prof. Yang Zhang.
I build multimodal virtual cell models: systems that learn how cells respond to drugs across morphology, transcriptomics, and molecular structure, together with the agent systems that turn such models into automated scientific discovery. My work has moved along one continuous line, from perceiving cell phenotypes in images, to aligning them with molecular readouts, to benchmarking and orchestrating virtual cell models end to end.
📫 Contact: Boom985426@gmail.com · WeChat: Boom_5426
I am always open to collaborations on virtual cell modeling, phenotypic drug discovery, and agentic systems for science. Feel free to reach out.
🔥 News
- 2026.06: 🇸🇬 Started a one-year visit to the School of Computing, National University of Singapore, hosted by Prof. Yang Zhang.
- 2026.05: 📄 CellScientist preprint released on arXiv (co-author).
- 2026.04: 🧬 MVCBench preprint released on bioRxiv.
- 2026.01: 🎉 One paper accepted at ICLR 2026 (co-author).
- 2025.12: 🎉 PhenoProfiler published in Nature Communications.
- 2025.08: 🎉 SpaIM published in Nature Communications.
📝 Selected Publications
Five representative works below. The complete list, including eight first-author journal papers, is on Google Scholar.
MVCBench: A Multimodal Benchmark for Drug-induced Virtual Cell Phenotypes
Bo Li, Qing Wang, Shihang Wang, Bob Zhang, Yuzhong Peng, Pinxian Zeng, Chengliang Liu, Mengran Li, Ziyang Tang, Xiaojun Yao, Chuxia Deng, Qianqian Song
bioRxiv 2026 · Preprint
TL;DR: A systematic benchmark of 24 drug-molecular and gene representation methods across ~1.1M drug-induced profiles. It exposes a modality-dependent asymmetry: advanced molecular representations substantially help morphological phenotype prediction but barely beat classical fingerprints for transcriptomic response, where task-specific gene representations outperform general-purpose foundation models.
PhenoProfiler: Advancing Phenotypic Learning for Image-based Drug Discovery
Bo Li, Bob Zhang, Chengyang Zhang, Minghao Zhou, Weiliang Huang, Shihang Wang, Qing Wang, Mengran Li, Yong Zhang, Qianqian Song
Nature Communications 17, 793 (2026) · Paper · Code · arXiv
TL;DR: The first end-to-end encoder for image-based phenotypic drug discovery. It replaces the conventional multi-step segmentation-and-feature-extraction pipeline with a single model, evaluated on ~400K high-content images and 8.42M single-cell images, improving accuracy and robustness by up to 20% over prior methods while cutting inference time by roughly 40×.
SpaIM: Single-cell Spatial Transcriptomics Imputation via Style Transfer
Bo Li, Ziyang Tang, Aishwarya Budhkar, Xiang Liu, Tonglin Zhang, Baijian Yang, Jing Su, Qianqian Song
Nature Communications 16, 7861 (2025) · Paper · Code
TL;DR: Recasts cross-modal imputation as style transfer, separating data-agnostic gene-expression “content” from platform-specific “style” to predict unmeasured genes in spatial transcriptomics from scRNA-seq. Across 53 datasets spanning sequencing- and imaging-based platforms, it consistently outperforms 12 state-of-the-art methods in gene coverage and expression accuracy.
Gene Expression Prediction from Histology Images via Hypergraph Neural Networks
Bo Li, Yong Zhang, Qing Wang, Chengyang Zhang, Mengran Li, Guangyu Wang, Qianqian Song
Briefings in Bioinformatics 25(6), bbae500 (2024) · Paper · Code
TL;DR: Builds a hypergraph over image patches using Euclidean distance and adjacent-position weighting, so that higher-order local correlations in whole-slide images can be exploited to predict spot-level gene expression.
Multi-scale Hypergraph-based Feature Alignment Network for Cell Localization
Bo Li, Yong Zhang, Chengyang Zhang, Xinglin Piao, Yongli Hu, Baocai Yin
Pattern Recognition 149, 110260 (2024) · Paper · Code
TL;DR: Reframes cell localization as a feature-alignment problem and introduces a multi-scale hypergraph module that adaptively aggregates multi-level features, substantially improving localization accuracy in dense tissue.
🛠 Open Source
Research code and community resources, 700+ GitHub stars in total.
| Project | What it is | Stars |
|---|---|---|
| Nature-Paper-Skills | Agent skills for drafting, revising, auditing, and resubmitting Nature-style manuscripts |  |
| Awesome-Virtual-Cell | Papers, datasets, benchmarks, and community resources for AI virtual cells |  |
| Awesome-Phenotypic-Drug-Discovery | Curated resources for phenotypic drug discovery |  |
| PhenoProfiler | End-to-end phenotypic profiling for image-based drug discovery |  |
| SpaIM | Style-transfer imputation for spatial transcriptomics |  |
| MHFAN | Multi-scale hypergraph feature alignment for cell localization |  |
| UM_CS_QE | Study resources for the University of Macau CIS Ph.D. qualifying exam |  |
📖 Education
-
2026.06 – 2027.06: National University of Singapore
- Visiting Student, School of Computing. Host: Prof. Yang Zhang
-
2024.08 – Present: University of Macau
- Ph.D. in Computer Science, Full Scholarship. Advisors: Prof. Bob Zhang, Prof. Qianqian Song
-
2021.09 – 2024.07: Beijing University of Technology
- M.Eng. in Electronic Information. Advisors: Prof. Yong Zhang, Prof. Baocai Yin
-
2017.09 – 2021.07: Beijing Information Science & Technology University
- B.Eng. in Robotics Engineering. Advisor: Prof. Hongbo Huang
🎖 Honors and Awards
- 2024: Ph.D. Scholarship, University of Macau
- 2024.07: Top 100 Graduates of BJUT (Top 100 / 6331)
- 2024.07 & 2021.07: Beijing Outstanding Graduate
- 2023.10: Xiaomi Scholarship
- 2023.10 & 2022.10: First-Class Academic Scholarship, BJUT
- 2022.10: National Scholarship
- 2021: First Prize, Science & Technology Innovation Scholarship, BISTU
- 2020.12: Second Prize, National Mathematics Competition
📜 Patents
Co-inventor of three Chinese invention patents on cell image density map generation and cell localization: CN115457546A, CN115810046A, CN115457547A.
💼 Academic Service
Journal reviewer: Science Advances, IEEE TIP, IEEE TNNLS, IEEE TCE, IEEE TSMCS, Medical Image Analysis, Briefings in Bioinformatics, BMC Biology, Engineering Applications of Artificial Intelligence, Expert Systems with Applications, Knowledge-Based Systems, CAAI Transactions on Intelligence Technology, and Artificial Intelligence Review.
Collaborations: Purdue University, Cornell University, University of Florida, National University of Singapore, Sun Yat-sen University, Sichuan University, Beijing University of Technology, and Macao Polytechnic University.
🧰 Skills
- Research: multimodal representation learning, medical image analysis, spatial omics modeling, phenotypic drug discovery, virtual cell modeling and benchmarking.
- Agent systems: multi-agent design, skill-based architectures, memory-augmented systems, reusable workflow design for scientific applications.
- Engineering: Python, PyTorch, CUDA, Linux.
