Skip to content

Repository files navigation

AgForce Enables Antigen-conditioned Generative Antibody Design

AgForce is a encoder-decoder architecture, that uses a graph neural network (GNN) as the encoder and specialized decoders for sequence-structure co-design. It applies framework dropout, gated bottlenecks, and hyperbolic cross attention that prevent the antibody shortcut path. In the decoder, a Mixture Density Network (MDN) sequence head with Potts-like pairwise coupling and annealed Multiple Choice Learning (aMCL) replaces the cross-entropy objective with a multi-component distribution whose optimal solution differs from the positional marginal. An antigen cycle consistency head routes gradients through the sequence decoder, forcing predicted distributions to encode antigen identity.

AgForce

Usage

# Quick test
python chimera_trainer.py --max_epoch 2 --no-wandb

# Full training
python chimera_trainer.py --split epitope_group --cdr_type 3

# Test only (load checkpoint)
python chimera_trainer.py --test_only --checkpoint /path/to/best.pt

About

AgForce generates realistic and diverse antigen-conditioned antibody CDRs via multiple choice learning

Resources

Stars

4 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages