TCR/CAR antagonism: data for theoretical modeling and results

Data and results related to mathematical modeling and theoretical analysis in the paper

Taisuke Kondo=, François X. P. Bourassa=, Sooraj R. Achar=, J.DuSold, P. F. Céspedes, M. Ando, A. Dwivedi, J. Moraly, C. Chien, S. Majdoul, A. L. Kenet, M. Wahlsten, A. Kvalvaag, E. Jenkins, S. P. Kim, C. M. Ade, Z. Yu, G. Gaud, M. Davila, P. Love, J. C. Yang, M. Dustin, Grégoire Altan-Bonnet, Paul François, and Naomi Taylor. "Engineering TCR-controlled Fuzzy Logic into CAR T-Cells Enhances Therapeutic Specificity", Cell [accepted in principle], 2025.(=: these authors contributed equally)

In that paper, our mathematical modeling efforts aimed to quantitatively predict TCR and CAR cross-receptor interactions in CAR T cells, and to interface with experimental data.  

This repository contains the data necessary to run the code provided on Github

https://github.com/frbourassa/tcr_car_antagonism

as well as several results and figures produced by that code, in particular MCMC simulation outputs, parameter estimates, and model predictions. 

Installation

Step 1: Download the code with the command

    git clone https://github.com/frbourassa/tcr_car_antagonism

Step 2: Download and unzip the data, results, and figures provided here, then place them in the main folder of the cloned code repository to replace the empty results/, data/, and figures/ folders. 

Step 3: Install the required Python packages (listed on our Github repository). 

Provided files

data.zip: data needed for analysis   ├── antagonism: measurements of TCR/TCR and TCR/CAR antagonism   ├── dose_response: dose response measurements, including public data from Luksza et al., Nature, 2022   ├── invivo: mouse survival and tumor progression measurements (Fig. 4)   └── surface_counts: calibration of receptor, MHC, and surface antigen abundances on different cell linesfigures.zip   ├── dose_response: analysis of some dose response data    │   ├── hhatv4_mcmc: dose response fits for the HHAT peptide library    │   └── mskcc_mcmc: dose response fits on the peptide libraries provided in Luksza et al.   ├── extra_predictions: model predictions in various conditions   ├── invivo: semi-quantitative model comparison with in vivo data   ├── mcmc_akpr_i: MCMC histograms and curve fits for TCR/TCR antagonism, revised AKPR model   ├── mcmc_akpr_i_withpriors: same, with alternate Gaussian priors (not included in the paper)   ├── mcmc_shp1: MCMC plots for TCR/TCR antagonism, classical AKPR model   ├── mcmc_tcr_car: MCMC plots for TCR/CAR antagonism   ├── mcmc_tcr_car_withpriors: same, with the alternate Gaussian priors   ├── mcmc_tcr_tcr_6f: MCMC plots for TCR/TCR antagonism in 4-ITAM (6F) T cells   ├── mcmc_tcr_tcr_6f_withpriors: same, with the alternate Gaussian priors   ├── model_analysis: theoretical analyses of the TCR/TCR and TCR/CAR models   ├── model_comparison: plots comparing the different models   ├── model_predictions: TCR/CAR antagonism predictions for T cells with different CAR and TCR ITAM numbers   └── model_predictions_withpriors: same, for the MCMC results with the alternate Gaussian priorsresults.zip    ├── for_plots: results saved by the code and used to generate final figures    ├── for_plots_withpriors: model confidence intervals for the MCMC runs with alternate priors    ├── mcmc: MCMC simulation results (chain samples) and analysis (best parameter fits for each k, m, f)    ├── mcmc_withpriors: same, for the alternate priors    └── pep_libs: results of the MCMC simulation analysis to fit curves on the dose response libraries 

License information

Data, results, and figures: CC-BY-4.0

The code itself, on Github, is licensed under the BSD-3-clause license.