Research Interests

My main research focuses on Statistical Theory and Methods and Statistical Genomics.

• In the realm of Statistical Theory and Methods, my research interests lie in the development of innovative methodologies pertaining to missing data analyisis, latent variable models and multi-response regression models. My goal is to address challenges related to variable type heterogeneity, nonlinear dependence, overdispersion, sparsity, individual spatial dependence, and supervised feature extraction. Through these efforts, I aim to enhance our understanding of complex statistical issues and pave the way for more accurate and effective statistical analysis.

• The Statistical Genomics I am studying falls under the umbrella of AI for Science. My primary focus lies in applying artificial intelligence (AI) techniques, such as Deep Learning and modern statistical machine learning, to scientific research and discovery, particularly in Genomics. This involves utilizing AI algorithms, machine learning, and data analytics to meticulously analyze vast amounts of scientific data, detect patterns and trends, make accurate predictions, and automate scientific processes. My specialization lies in developing computational algorithms and statistical methods aimed at addressing frontier issues in genomics, including resolving tissue heterogeneity, pinpointing marker/signature genes, refining expression patterns, and integrating multi-section and multi-modality data.

• Statistical Theory and Methods

  • Vector/matrix/tensor data analysis via latent variable models

  • Missing data analysis

  • Multi-repsonse regression analysis

• Statistical Genomics

  • Tissue heterogeneity analysis

  • Multi-section/modality integration analysis

  • Cell-gene relationship modelling

  • Dissecting mechnisms in disease (i.e., tumour immunology)

Grants

• No.12371283, National Natural Science Foundation of China, 1/2024-12/2027
  • Title: Statistical methods for dimensionality reduction, automatic annotation, data integration, and spatial deconvolution in spatially resolved transcriptomics
  • Role: participant
• No.12071372, National Natural Science Foundation of China, 1/2021-12/2024
  • Title: Statistical inference and manifold structure research based on F-divergence for degenerate data
  • Role: participant

• MOE2018-T2-2-006 Ministry of Education, Singapore, 10/2019-04/2022

  • Title: Methods to identifying genetically regulated components using transcriptome data
  • Role: participant

• No.11931014, National Natural Science Foundation of China, 1/2020-12/2024

  • Title: Semiparametric integrative regression inference
  • Role: participant

• No.113610004007000040, Excellent Doctoral Dissertation Project, SWUFE, 1/2020 - 12/2020

  • Title: Theory and Application of High-dimensional Latent Factor Models
  • Role: principal investigator
• No.11571282, National Natural Science Foundation of China, 1/2016-12/2019
  • Title: Clustering analysis for trajectory data
  • Role: participant

Patents

• Jin Liu, Wei Liu; Transcriptome Data Integration, 2022-06-02, Singapore, E202206020334XPF1WK, Application No.: 10202250010J