For MHC-I protein-peptide binding predictions: Deep Learning model with CNN and Snakemake workflow.
This repository includes a Tensorflow based Deep Learning model, benchmark datasets and a Snakemake workflow for MHC-I protein-peptide binding predictions. The workflow allows running all the models for each test allele at once in parallel over clusters.
The Tutorial.txt file provides instructions for running the current model.
This repository is a supporting material for the following manuscript and all the details can be seen in it: G. Altay, "Tensorflow Based Deep Learning Model and Snakemake Workflow for Peptide-Protein Binding Predictions", bioRxiv, 2018. https://www.biorxiv.org/content/early/2018/09/18/410928
"This software is provided with absolutely no warranty."