Timezone: »

 
AbODE: Ab initio antibody design using conjoined ODEs
Yogesh Verma · Markus Heinonen · Vikas K Garg

@
in Structured Probabilistic Inference and Generative Modeling »
Event URL: https://openreview.net/forum?id=eMiAK9nMfe »

Antibodies are Y-shaped proteins that neutralize pathogens and constitute the core of our adaptive immune system. De novo generation of new antibodies that target specific antigens holds the key to accelerating vaccine discovery. However, this co-design of the amino acid sequence and the 3D structure subsumes and accentuates, some central challenges from multiple tasks including protein folding, inverse folding, and docking. We strive to surmount these challenges with a new generative model AbODE that extends graph PDEs to accommodate both contextual information and external interactions. Unlike existing approaches, AbODE uses a single round of full-shot decoding, and elicits continuous differential attention that encapsulates, and evolves with, latent interactions within the antibody as well as those involving the antigen. We unravel fundamental connections between AbODE and temporal networks as well as graph-matching networks. The proposed model significantly outperforms existing methods on standard metrics across benchmarks.

Keywords: Antibody Design · Neural ODEs

Author Information

Yogesh Verma (Aalto University)

I’m a Doctoral Researcher at (Qurious) Quantum Machine Learners, Aalto University, working with Prof. Vikas Garg, Prof. Samuel Kaski and Dr. Markus Heinonen. I’m interested in physics inspired geometric deep learning, AI4Science and machine learning for molecules and drug design, and many more thing

Markus Heinonen (Aalto University)
Vikas K Garg (Massachusetts Institute of Technology)

More from the Same Authors