Mengji Zhang, PI, Assistant Professor at Department of Biomedical Engineering, Southern University of Science and Technology (SUSTech). Dr. Zhang received her Ph.D. from the Department of Biomedical Engineering at Shanghai Jiao Tong University. She subsequently worked as a postdoctoral fellow at Yale School of Medicine and as a Senior Postdoctoral Fellow at the Wellcome Sanger Institute, Cambridge. She joined the Department of Biomedical Engineering at SUSTech as a full-time faculty member in September 2026. Her work has been published in leading international journals including Nature Biomedical Engineering, npj Systems Biology and Applications, Advanced Science, Advanced Functional Materials, PNAS, Advanced Materials, Angewandte Chemie International Edition, and IEEE Transactions on Industrial Electronics.

The Zhang Lab welcomes applications on a rolling basis from undergraduate students, master's students, Ph.D. students, postdoctoral researchers, research assistants, and visiting scholars. Interested applicants should send their CV and relevant materials by email. Prospective Ph.D. students are required to first work as a visiting student or research assistant, and will be admitted only after an evaluation period.

Homepage: https://mjdelta.github.io

Education

2018–2023 Ph.D., Department of Biomedical Engineering, Shanghai Jiao Tong University

2014–2018 B.S., Information Management and Information Systems, Dongbei University of Finance and Economics

Professional Experience

2026–present Assistant Professor, Department of Biomedical Engineering, Southern University of Science and Technology

2025–2026 Senior Postdoctoral Fellow, Wellcome Sanger Institute, Cambridge

2024–2025 Postdoctoral Fellow, Yale University

Research Directions

Multiscale Generative AI for Life Sciences: From DNA Sequences to Cellular Microenvironments and Dynamic Phenotypes

Our research focuses on building generative AI models that span DNA sequence, molecular, cellular, and systems scales, connecting the full pipeline from genomic sequence to cellular microenvironment to dynamic phenotype, to advance mechanistic discovery and clinical translation in the life sciences. We focus on three specific directions:

1. Regulatory Genomics and Sequence-to-Function AI: Starting from DNA sequence, we build sequence-to-function foundation models for deciphering genomic regulation, revealing the causal relationships between sequence variation and gene regulation, providing a computational foundation for interpreting disease risk loci and enabling precision diagnostics.

2. Spatial and Multimodal Omics AI: Addressing the complex interactions between cells and their microenvironments, we develop generative representation learning methods that integrate multimodal data such as spatial transcriptomics, proteomics, and metabolomics, characterizing cell states and microenvironmental regulatory patterns within a unified latent space, and enabling systematic modeling of cell-cell communication and tissue microenvironments.

3. Dynamic Phenotypes and Neurocomputational AI: Addressing dynamic biological processes such as neural activity and disease progression, we build generative frameworks capable of modeling state evolution and trajectory prediction, combined with perturbation modeling to simulate cellular/disease state transitions and design interventions, serving risk assessment and personalized treatment design.

Together, these three directions support a technical framework of "data representation—generative modeling—perturbation simulation," ultimately serving two major goals: mechanism discovery (genomic regulation, cellular microenvironmental interaction, cell/disease trajectory analysis) and clinical translation (precision diagnostics, prognostic evaluation, personalized intervention).

Selected Publications

Zhang, M.et al. A self-adaptive and versatile tool for eliminating multiple undesirable variations from transcriptome. Nature Biomedical Engineering, 2026, 10, 413–426. (Cover article)

Zhang, M.* et al. A deep position-encoding model for predicting olfactory perception from molecular structures and electrostatics. npj Systems Biology and Applications, 2024, 10, 76.

Zhang, M. et al. Ultra-fast label-free serum metabolic diagnosis of coronary heart disease via a deep stabilizer. Advanced Science, 2021, 8, 2101333.