Antibody affinity modeling

This repository contains examples of how to work with AI models for protein design.

• esmfold_multimer.ipynb demonstrates how to predict the folding of an antibody sequence using HuggingFace's ESMFold model. This model can be used to predict the folding on any protein.

• antibody-affinity.ipynb demonstrates how to load the antibody affinity dataset from TDCommons and train a neural network with PyTorch Lightning. (Currently the model makes very bad predictions.)

A 3D model of an Immunoglobulin molecule, showing heavy chains in blue and light chains in green. By the National Library of Medicine. Public domain.

Installation

1. Install miniconda
2. conda env create -f environment.yml to create the conda environment called antibody-env.
3. conda activate antibody-env to activate the conda environment.
4. jupyter lab to run a Jupyter Lab server. Click on the url in the output log.

Jupyter Server 2.13.0 is running at:
    http://localhost:8888/lab?token=55d9dont7d55usea9fc266notgooda5d91fakeda8ae50783caedd71

5. Run the notebooks.