GENTLE: a novel bioinformatics tool for generating features and building classifiers from T cell repertoire cancer data

Abstract

Background: In the global effort to discover biomarkers for cancer prognosis, prediction tools have become essential resources. TCR (T cell receptor) repertoires contain important features that differentiate healthy controls from cancer patients or differentiate outcomes for patients being treated with different drugs. Considering, tools that can easily and quickly generate and identify important features out of TCR repertoire data and build accurate classifiers to predict future outcomes are essential.

Results: This paper introduces GENTLE (GENerator of T cell receptor repertoire features for machine LEarning): an open-source, user-friendly web-application tool that allows TCR repertoire researchers to discover important features; to create classifier models and evaluate them with metrics; and to quickly generate visualizations for data interpretations. We performed a case study with repertoires of TRegs (regulatory T cells) and TConvs (conventional T cells) from healthy controls versus patients with breast cancer. We showed that diversity features were able to distinguish between the groups. Moreover, the classifiers built with these features could correctly classify samples ('Healthy' or 'Breast Cancer')from the TRegs repertoire when trained with the TConvs repertoire, and from the TConvs repertoire when trained with the TRegs repertoire.

Conclusion: The paper walks through installing and using GENTLE and presents a case study and results to demonstrate the application's utility. GENTLE is geared towards any researcher working with TCR repertoire data and aims to discover predictive features from these data and build accurate classifiers. GENTLE is available on https://github.com/dhiego22/gentle and https://share.streamlit.io/dhiego22/gentle/main/gentle.py .

Keywords: Feature selection; Machine learning tools; T cell receptor repertoire.

Fig. 1

GENTLE Workflow

Fig. 2

A–D Scatter Plot of each dimension using the TConvs dataset. E–H Scatter Plot of each fdimension using the TRegs dataset. It is important to emphasize that we used the most predictive features on each scatter plot, according to the feature selection methods

Fig. 3

A–C Stratified validation using threefold and 100 repeats of the classifiers trained with the TConvs dataset where A is the Gaussian Naive Bayes, B is the Linear Discriminant Analysis and C are Linear Regression and Decision Tree classifiers. D Confusion matrix of the model trained with the TConvs dataset (train) and validated with the TRegs dataset (test). E Stratified validation using threefold and 100 repeats of the Decision Tree classifier trained with the TRegs dataset. F Confusion matrix of the model trained with the TRegs dataset (train) and validated with the TConvs dataset (test)