Prediction of tumor-reactive T cell receptors from scRNA-seq data for personalized T cell therapy

Prediction of tumor-reactive T cell receptors from scRNA-seq data for personalized T cell therapy

07 March 2024 | C. L. Tan, K. Lindner, T. Boschert, Z. Meng, A. Rodriguez Ehrenfried, A. De Roia, G. Haltenhof, A. Faenza, F. Imperatore, L. Bunse, J. M. Lindner, R. P. Harbottle, M. Ratliff, R. Offringa, I. Poschke, M. Platten & E. W. Green
The study presents a machine learning classifier called *predicTCR* that identifies tumor-reactive T cell receptors (TCRs) from single-cell RNA and VDJ sequencing (scRNA + VDJ-seq) data. This approach is designed to overcome the limitations of current methods, which rely on gene set enrichment analysis and can be time-consuming and costly. *predicTCR* combines high-throughput TCR cloning and reactivity validation to train a classifier that can identify individual tumor-reactive TILs in an antigen-agnostic manner. The classifier outperforms previous gene set enrichment-based approaches, increasing specificity and sensitivity from 0.38 to 0.74. By predicting tumor-reactive TCRs in a matter of days, *predicTCR* can prioritize TCR clonotypes for personalized T cell therapy, accelerating the manufacturing process. The study demonstrates the effectiveness of *predicTCR* through experimental validation and benchmarking against other prediction methods, showing that it can accurately identify tumor-reactive TCRs in diverse cancer types.The study presents a machine learning classifier called *predicTCR* that identifies tumor-reactive T cell receptors (TCRs) from single-cell RNA and VDJ sequencing (scRNA + VDJ-seq) data. This approach is designed to overcome the limitations of current methods, which rely on gene set enrichment analysis and can be time-consuming and costly. *predicTCR* combines high-throughput TCR cloning and reactivity validation to train a classifier that can identify individual tumor-reactive TILs in an antigen-agnostic manner. The classifier outperforms previous gene set enrichment-based approaches, increasing specificity and sensitivity from 0.38 to 0.74. By predicting tumor-reactive TCRs in a matter of days, *predicTCR* can prioritize TCR clonotypes for personalized T cell therapy, accelerating the manufacturing process. The study demonstrates the effectiveness of *predicTCR* through experimental validation and benchmarking against other prediction methods, showing that it can accurately identify tumor-reactive TCRs in diverse cancer types.
Reach us at info@study.space
[slides and audio] Prediction of tumor-reactive T cell receptors from scRNA-seq data for personalized T cell therapy