Any tools for predicting binding affinity between TCR and epitode?

Question

[CY]

I am working on a immunotherapy related project that identify tumor epitode (neiantigen) and TCR-seq. Say I already have both epitode and sequences of T/B cell CDR3 region available. Is there any tools available for predicting their binding affinity?

Answer 1

[niuyw]

Hi! I found these tools by Google, but I haven't try any of them. Hope to be helpful.

• OncoPeptVAC: a robust TCR binding algorithm to prioritize neoepitope using tumor mutation (DNAseq) and gene expression (RNAseq) data.
• TCRex: a webtool for the prediction of T-cell receptor sequence epitope specificity.
• NetTCR: sequence-based prediction of TCR binding to peptide-MHC complexes using convolutional neural networks.
• Repitope: epitope immunogenicity prediction through in silico TCR-peptide contact potential profiling.

Answer 2

[igor]

I am not too familiar with this type of analysis, but IEDB Analysis Resource may be what you are looking for:

Welcome to the Immune Epitope Database Analysis Resource. This site provides a collection of tools for the prediction and analysis of immune epitopes. It serves as a companion site to the Immune Epitope Database (IEDB) , a manually curated database of experimentally characterized immune epitopes.

[CY]

I believe that IEDB predicts MHC - epitope binding. Not epitope - TCR binding which is the next step once the epitope got presented to the cell surface.

[igor]

Thanks for clarifying.

IEDB has a lot of different tools, so I assumed they covered the full range. Maybe not.

Answer 3

[tim]

Hi! If you still need help with your problem, one of the leading tools to determine the epitope specificity of your T-cell receptor sequences is IMW-DETECT. If you want to know how powerful those bindings are. We don't have that solution, but by running our tool, we can help to filter out less specific TCRs or false-positive TCRs. We are running a free access through immunewatch.com/detect-early-access-program/. Disclosure, I am working on the product.