Author: Lai Wenpu

Publisher: School of Medicine

April 16, 2025

A research team led by Luo Junhong from the School of Medicine at Jinan University has made significant strides in the analysis of T cell functionality, as detailed in their newly published paper titled MIST: An interpretable and flexible deep learning framework for single-T cell transcriptome and receptor analysis in Science Advances.

(Screenshot of the paper)

Key Innovations

The study introduces MIST (Multi Insight for T cell), an artificial intelligence model built upon the Variational Autoencoder deep learning framework. This innovative model is designed for joint analysis of single-cell RNA sequencing (scRNA-seq) and single-cell T cell receptor sequencing (scTCR-seq) data. MIST addresses the limitations found in existing methodologies, offering a robust analytical framework that integrates multiple omics data types with enhanced interpretability and adaptability.

Abstract Highlights

The paper emphasizes the impact of analyzing transcriptomic and TCR features at the single-cell level, which enables deeper research into T cell immune functions. Key features of MIST include:

•Three Latent Spaces: MIST operates within three distinct latent spaces: gene expression, TCR, and a joint latent space. This structured approach allows for comprehensive integration of data.

•Interpretability and Flexibility: The model effectively resolves key T cell characteristics such as functional status, clone amplification patterns, and antigen specificity, facilitating a more nuanced understanding of T cell behavior.

•Experimental Insights: Utilizing MIST, the research team investigated antigen-specific T cells and datasets related to lung cancer immunotherapy and COVID-19. They uncovered critical insights into the heterogeneity of CXCL13+ subsets in lung cancer-infiltrating CD8+ T cells and their relationship with immunotherapy, particularly regarding anti–PD-1 therapy, revealing patterns not previously reported.

Contributions and Support

Professor Luo Junhong serves as the independent corresponding author of the paper, while Dr. Lai Wenpu of the First Affiliated Hospital of Jinan University is the independent first author. Dr. Lai's doctoral studies were guided by Researcher Li Yangqiu and Professor Luo Junhong, part of a unique Clinical Medicine+X doctoral program.

This significant research was supported by the National Natural Science Foundation of China Interdisciplinary Research Program and the Central University Basic Research Funds Interdisciplinary Cultivation Special Project of Jinan University.

For further reading, the full paper is available:

https://www.science.org/doi/10.1126/sciadv.adr7134