We aim to develop mass spectrometry-based protein technologies to decipher the proteome complexity, with a focus on understudied proteins, and facilitate disease diagnosis, treatment and drug discovery. We are developing the following proteomic technologies.

PCT-DIA: Through the utilization of pressure-cycling technology in conjunction with Data-Independent Acquisition (DIA) mass spectrometry, we developed practical high-throughput proteomic analysis of biopsy tissue specimens (Guo et al., Nature Medicine, 2015; Zhu, et al, Mol Oncology, 2019; Cai et al., Nature Protocols, 2022). This technology allowed us to report the first multi-organ proteome of COVID-19 autopsies, and uncovered multiple potential drug targets (Cell, 2021) during the pandemic.

OmniProt: We have also developed OmniProt beads, in collaboration with Westlake Omics, to allow high-throughput and deep proteomic analysis of blood samples.

ProteomEx and FAXP: We developed expansion proteomics techniques (Li et al., Nature Communications, 2022; Dong, 2024, submitted), which analyze spatial and single cell proteomics by physical expansion of formalin-fixed tissue sections. Additionally, we are actively delving into subcellular proteomics.

Perturbation proteomics: we propose an emerging area of perturbation proteomics. We are building up robotics systems, acquiring perturbing 10,000s proteomes with drug or genetic perturbations. We collaborate with computer scientists in building AI models for predicting protein functions, drug actions, and drug synergies.

Proteomic database: We are building a proteomic big dataset called ProteinTalks: db.prottalks.com.

Large AI models: Together with Westlake Omics, we are condensing information hidden in billions of mass spectra into MS-BERT and MS-GPT models for analyze mass spectrometry-based spectra.

Robotics for proteomics: we are building a series robotic system to automate proteomic sample preparation.

In the meantime, we are applying these technologies to address the following biomedical questions.

Disease diagnosis and stratification: We developed AI-empowered protein-based biomarkers for stratifying COVID-19 (Shen, et al, Cell, 2020; Bi, et al, Cell Reports, 2022), metabolic diseases (Cai, et al, Cell Reports Med, 2023), and thyroid nodules (Sun, et al. Cell Discovery, 2022; Cai, et al, submitted). Ongoing biomarker studies in our lab focuses on Alzheimer’s Diseases and lung cancers. We are collaborating with Westlake Omics to translate the protein biomarkers established jointly by Guomics and clinicians into clinical assays. Prediction of protein function and drug action: We are building computer models for predicting protein function, with a focus on understudied proteins, and drug actions. This is in collaboration with colleagues in the Grand Challenge initiative (HUPO), and the Understudied Protein Initiative (UPI). Aging brain: Growing research interests in our lab are being directed towards aging brain. We are collaborating with neurobiologists and clinicians in understanding the protein perturbations in aging brains. A special focus is the Alzheimer’s disease. Proteome complexity science: Life activities are inherently complex. Artificial reduction of the complexity is the dominant strategy in biomedical research, leading to numerous breakthroughs in life sciences and medicine. Currently, we are facing complex systems that can be hardly tackled with human brains. What are the mysteries beneath the protein dynamics?

欢迎有意者附上研究兴趣与 PI 交流。