Shuguang Wang
About Publications Research Interests CV
Background

About Me

I am a Ph.D. candidate in Bioinformatics at Tongji University, graduating in June 2026. My research sits at the intersection of AI and single-cell biology, with a focus on systematic benchmarking, deep generative modeling, and multi-omics data integration.

My doctoral work includes two co-first-author publications in Nature Methods — large-scale evaluation studies that defined gold standards for single-cell multi-modal integration and perturbation response prediction. I also contributed to developing deep generative models for multi-omics profiling (Genome Biology), and applied spatial transcriptomics to uncover the regulatory role of CCN1 in microglia spatial organization (Genes & Diseases).

Selected Work

Publications

Nature Methods2025
Benchmarking Single-Cell Multi-Modal Data Integrations
Fu, S.*, Wang, S.*, Si, D., Li, G., Gao, Y., Liu, Q.
Paper → GitHub →
Nature Methods2025
Benchmarking Algorithms for Generalizable Single-Cell Perturbation Response Prediction
Wei, Z.*, Wang, Y.*, Gao, Y.*, Wang, S.*, ..., Liang, A., Chuai, G., Liu, Q.
Paper → GitHub →
Genome Biology2022
A Deep Generative Model for Multi-View Profiling of Single-Cell RNA-seq and ATAC-seq Data
Li, G., Fu, S., Wang, S., Zhu, C., Duan, B., Tang, C., Chen, X., Chuai, G., Wang, P., Liu, Q.
Paper → GitHub →
Genes & Diseases2025
Revealed the Regulatory Role of CCN1 to Microglia Distribution through Region-Specific Cellular Interactions
Bai, Z.*, Chang, Z.*, Wang, S., ..., Shen, Q., Gao, S., Gao, Y.
Paper →
In Preparation
Systematic Evaluation of RNA Velocity and Spatiotemporal Dynamics Inference Methods in Single-Cell Biology
Wang, S., ..., Liu, Q.
Looking Forward

Research Interests

Spatiotemporal Virtual Cell Modeling

Integrating multi-omics data with foundation models to build dynamic, multi-dimensional virtual cell systems capable of simulating cellular behavior over space and time — enabling in silico perturbation discovery and predictive modeling of cell fate.

Multi-Modal Biomedical Data Integration

Bridging multi-omics biological data with clinical and medical imaging modalities to create unified analytical frameworks — translating molecular insights into actionable knowledge for precision medicine and therapeutic development.

Defining Problems and Gold Standards for Computational Biology

Identifying fundamental biological questions that can be rigorously formulated as computational challenges, and establishing principled evaluation frameworks and gold-standard benchmarks to guide the development of next-generation methods.