AI-Driven Digital Twin of Cell State Evolution and Perturbation

The advent of high-throughput omics technologies has enabled a more quantitative and integrative understanding of cellular dynamics. Our research focuses on developing AI-driven digital twins of cells, creating dynamic computational models that simulate cell state evolution and their perturbation in development, disease and therapeutic intervention. By integrating multi-omics data with deep learning, we aim to delucidate the transcriptional logic and regulatory networks that govern cell fate, moving from static observation to dynamic, predictive simulation of cellular processes and transitions.

Major research directions:

1. Integrating large-scale multi-omics data to develop deep learning models and algorithms for characterizing and interpreting cellular identities and states.
2. Building digital twin models of cells to simulate and predict the dynamic evolution of cellular states during development, disease progression, and therapeutic intervention.
3. Reconstructing the trajectories of disease initiation and progression to identify critical transition points and therapeutic windows, enabling early warning and precise interception of disease.

To achieve these goals, we use a wide range of cutting-edge technologies, including deep learning, bioinformatics, genomics, CRISPR gene editing, high-throughput screening, and next generation sequencing techniques (such as STARR-Seq、Hi-C、ATAC-Seq、ChIP-Seq、RNA-Seq and so on).