A new human T cell annotation tool and reference database

T cells are essential components of the immune system and play critical roles in immune responses against infections, tumors, and autoimmune diseases. T cells are highly heterogeneous, with distinct subtypes and states serving various functions. Therefore, a comprehensive and accurate annotation of T cell subtypes and states in various diseases and tissues is essential for understanding the function of T cells and the immune environment. However, there are no references and annotation tools for T cell subtypes and states.

Now, researchers from West China Hospital of Sichuan University have constructed a comprehensive human T cell reference for 68 subtypes and states, alongside STCAT (Single T Cell Annotation Tool), an automated annotation tool achieving 28% higher accuracy than existing methods. The authors also develop a TCellAtlas database, which allows users to browse T cell expression profiles and analyze customized scRNA-seq data by STCAT. This study was published in Science Bulletin.

Researchers collected more than 1.3 million high-quality T cells across 35 conditions and 16 tissues and used a two-level annotation process to accurately annotate the T cell subtypes and states. Based on this annotation, researchers constructed a reference for T cell subtypes and states, which included 68 T cell subtypes and states.

Based on the high-confidence T cell subtypes and states reference, they developed the STCAT tool. STCAT can automatically annotate T cell subtypes and states from the scRNA-seq data of different conditions and tissues. The accuracy of STCAT was 28% higher than the existing tools validated on six independent datasets, including cancer and healthy samples.

Using STCAT, researchers can integrate and compare the results of different datasets. As applications, they consistently discovered that CD4⁺ Th17 cells were enriched in late-stage lung cancer patients in multiple datasets, whereas mucosal-associated invariant T cells were prevalent in milder-stage COVID-19 patients. Researchers also confirmed a decrease in Treg cytotoxicity in post-treatment ovarian cancer and CD8⁺ Tn IFN-response cells were enriched in post-immunotherapy nonresponse samples.

Systematic landscape analyses of CD4⁺ and CD8⁺ T cell references revealed that CD4⁺ regulatory T (Treg) cells were enriched in tumor samples and that CD8⁺ naive-related cells were abundant in healthy individuals. Researchers also explored the molecular features of Treg cells with different states, including classic, activated, cytotoxicity, IFN-response, naive-like, and stress-response. Researchers have found differences in gene expression, transcription factors, metabolism, gene enrichment, disease distribution, and cytokine responsiveness among Treg cells in different states. Finally, researchers deposited all the references of T cell subtypes and their expression markers into a TCellAtlas database, which allows users to browse T cell expression profiles and analyze customized scRNA-seq data by STCAT.

The source code and manual for STCAT are freely available at https://github.com/GuoBioinfoLab/STCAT. The T cell scRNA-seq database is available at https://guolab.wchscu.cn/TCellAtlas.

The West China Biomedical Big Data Center at West China Hospital of Sichuan University integrates multidisciplinary medical expertise, vast clinical data from four affiliated hospitals, and cross-disciplinary research to advance innovations in health management, precision medicine, and disease prevention.

---

---