I study computational models that connect molecular and cellular perturbations with phenotypic outcomes. My current interests include perturbation biology, AI virtual cells, single-cell transcriptomics, AI-driven drug discovery, and phenotype-based drug design.
PhD in Biology
Tsinghua University, 2025-present
Advisor: Prof. Zemin Zhang
B.S. in Biomedical Science
Shanghai Jiao Tong University, Zhiyuan Honors Program, 2021-2025
Molecular dynamics, docking, homology modeling, pathway analysis, Markov state modeling, and allosteric mechanism analysis.
Single-cell foundation model pretraining, deployment, post-training, perturbation response modeling, and phenotype prediction.
Transformer-based, flow matching-based, diffusion-based, and agent-based models, with biology-prior embeddings for biomedical representation learning.
Modeling how genetic, molecular, and environmental perturbations reshape cell states, with emphasis on deployable single-cell foundation models and post-training for perturbation response prediction.
Using single-cell data to characterize tumor microenvironment structure, cellular state transitions, and phenotype-level readouts for computational biology workflows.
Connecting molecular generation, target-aware evaluation, and cellular phenotypes so that drug design is guided by both mechanism and response.
Studying conformational dynamics, docking, Markov state models, and allosteric communication in CRISPR-Cas systems and drug-discovery settings.
Ongoing work on bidirectional genomic representation learning. The project is being actively advanced after the KDD submission cycle and is presented here as a manuscript in …
Built an end-to-end AI workflow that generated 3.7 million TCM-like molecules, supports target-specific generation, and links molecular generation to TCM-like quality control and …
Combined computation and experiments to map remote allosteric inhibition of SauCas9 by AcrIIA14, highlighting dynamic inhibitory states and potential routes for allosteric Cas9 …
PhD-stage research on tumor microenvironment, perturbation biology, and single-cell modeling.
Developed TCM-Navigator, an AI-assisted workflow for TCM-like molecular generation, quality control, and target-aware evaluation.
Studied CRISPR-Cas9 allostery, conformational dynamics, and allosteric database resources for mechanism-aware drug discovery.