NetTCR - 2.1
Sequence-based prediction of peptide-TCR binding.
NetTCR-2.1 predicts binding probability between a T-cell receptor (TCR) CDR loops and MHC-I peptides
Instructions for NetTCR-2.1
- The server only accepts amino acid sequences takes in newspace separated TCR CDR sequences. The CDR sequences should be comma-separated. (Load Example on Submission page for illustration of the format);
- The sequences should be maximum 30 amino acid long and should contain only uppercase standard amino acid;
- Paste CDR sequence(s) into the box, or load an example file, or load a file from your lcoal machine. In case CDR3 is selected, two columns are expexted in the input file; for "All CDRs" option, 6 columns are expexted. The input file should be a text or .csv file with no headers for the columns.
- Select the desire CDR3s to use;
- Select the peptide(s) to pair the CDR sequences with.
Prediction of T cell receptor (TCR) interactions with MHC-peptide complexes remains highly challenging. This challenge is primarily due to three dominant factors: data accuracy, data scarceness, and problem complexity.
Here, we showcase that "shallow" convolutional neural network (CNN) architectures are adequate to deal with the problem complexity imposed by the length variations of TCRs.
We demonstrate that current public bulk CDR3β-pMHC binding data overall is of low quality and that the development of accurate prediction models is contingent on paired α/β TCR sequence data corresponding to at least 150 distinct pairs for each investigated pMHC.
In comparison, models trained on CDR3α or CDR3β data demonstrated a variable and pMHC specific relative performance drop. Together these findings support that T cell specificity is predictable given the availability of accurate and sufficient paired TCR sequence data.