maxhebditch/abpred repository overview

⁠Introduction

This repository contains all of the code required to run the protein-sol abpred⁠ machine learning models trained on the Jain 2017 dataset⁠ of 137 clinical stage therapeutic antibodies sequences and performance on 12 biophysical characterisation platforms.

The paper describing this work has been submitted for consideration and the preprint will be available soon.

This code can run predictions on multiple proteins.

Code is also available on Github⁠

⁠Running this image

1. docker pull maxhebditch/abpred:latest
2. Name the input fasta file as abpred.fasta in current directory
3. Run docker run --rm -v $(pwd)/:/abpred/host maxhebditch/abpred
4. Wait for output
5. Predictions from machine learning models will be in a directory called abpred_outputs in your current directory.