T-cells play a key role in adaptive immunity by mounting specific responses against diverse pathogens. An effective binding between T-cell receptors (TCRs) and pathogen-derived peptides presented on Major Histocompatibility Complexes (MHCs) mediate an immune response. However, predicting these interactions remains challenging due to limited functional data on T-cell reactivities. Here, we introduce a computational approach to predict TCR interactions with peptides presented on MHC class I alleles, and to design novel immunogenic peptides for specified TCR-MHC complexes. Our method leverages HERMES, a structure-based, physics-guided machine learning model trained on the protein universe to predict amino acid preferences based on local structural environments. Despite no direct training on TCR-pMHC data, the implicit physical reasoning in HERMES enables us to make accurate predictions of both TCR-pMHC binding affinities and T-cell activities across diverse viral epitopes and cancer neoantigens, achieving up to 72% correlation with experimental data. Leveraging our TCR recognition model, we develop a computational protocol for {de novo} design of immunogenic peptides. Through experimental validation in three TCR-MHC systems targeting viral and cancer peptides, we demonstrate that our designs---with up to five substitutions from the native sequence---activate T-cells at success rates of up to 50%. Lastly, we use our generative framework to quantify the diversity of the peptide recognition landscape for various TCR-MHC complexes, offering key insights into T-cell specificity in both humans and mice. Our approach provides a platform for immunogenic peptide and neoantigen design, opening new computational paths for T-cell vaccine development against viruses and cancer.
This repository contains data and code to generate the results in the paper "T-cell receptor specificity landscape revealed through de novo peptide design" by Visani G.M. et al.
The code is contingent upon the installation of the following three tools:
- HERMES (https://github.com/StatPhysBio/hermes)
- ProteinMPNN (fork with code to score and design peptides https://github.com/gvisani/ProteinMPNN-copy)
- TCRdock (fork with code to score peptides https://github.com/gvisani/TCRdock-copy)
Code and benchmarking results of TCR-pMHC binding affinity prediction across different peptides can be found in mutation_effects/nyeso and mutation_effects/tax.
Code and benchmarking results of T-cell activity prediction across different peptides can be found in mutation_effects/mskcc and mutation_effects/hsiue_et_al.
Code for running HERMES-relaxed to score peptides can be found in mutation_effects/src.
If you're only looking for our HERMES-made peptide designs with in-vitro T-cell activity measurements, you can find them all in peptide_designs/All-designs.xlsx.
Scripts to generate peptides with different algorithms, benchmarking results (via TCRdock PAE), and post-in-vitro-experiment analysis can be found in peptide_designs/nyeso, peptide_designs/ebv, peptide_designs/magea3_and_titin.
Code for using HERMES-relaxed to design peptides can be found in peptide_designs/src. What is currently present is an older version of the code that is not well integrated with the HERMES repository, we will provide updated code soon.
We use the entropy of HERMES-fixed's predicted Position Weight Matrix of peptides to quantify the diversity of the TCR recognition landscape. Specifically, we compute peptide entropies conditioned on TCR-MHC structures using HERMES-fixed, and compare them to the entropies of the same peptides conditioned on the MHC only, grabbed from the MHC Motif Atlas (http://mhcmotifatlas.org/home).
Code and data can be found in tcr_specificity.