@article{4b7000545c2b46cba5680f07aea1804b,
title = "Antibody structure prediction using interpretable deep learning",
abstract = "Therapeutic antibodies make up a rapidly growing segment of the biologics market. However, rational design of antibodies is hindered by reliance on experimental methods for determining antibody structures. Here, we present DeepAb, a deep learning method for predicting accurate antibody FV structures from sequence. We evaluate DeepAb on a set of structurally diverse, therapeutically relevant antibodies and find that our method consistently outperforms the leading alternatives. Previous deep learning methods have operated as “black boxes” and offered few insights into their predictions. By introducing a directly interpretable attention mechanism, we show our network attends to physically important residue pairs (e.g., proximal aromatics and key hydrogen bonding interactions). Finally, we present a novel mutant scoring metric derived from network confidence and show that for a particular antibody, all eight of the top-ranked mutations improve binding affinity. This model will be useful for a broad range of antibody prediction and design tasks.",
keywords = "DSML 3: Development/pre-production: Data science output has been rolled out/validated across multiple domains/problems, antibody design, deep learning, model interpretability, protein structure prediction",
author = "Ruffolo, {Jeffrey A.} and Jeremias Sulam and Gray, {Jeffrey J.}",
note = "Funding Information: We thank Dr. Sai Pooja Mahajan for helpful discussions and advice. This work was supported by National Institutes of Health grants R01-GM078221 and T32-GM008403 (J.A.R.) and AstraZeneca (J.A.R.). Computational resources were provided by the Maryland Advanced Research Computing Cluster (MARCC). Funding Information: We thank Dr. Sai Pooja Mahajan for helpful discussions and advice. This work was supported by National Institutes of Health grants R01-GM078221 and T32-GM008403 (J.A.R.) and AstraZeneca (J.A.R.). Computational resources were provided by the Maryland Advanced Research Computing Cluster (MARCC). J.A.R. and J.J.G. conceptualized the project. All authors developed the methodology. J.A.R. developed the software and conducted the investigation. J.S. and J.J.G. supervised the project. All authors wrote the manuscript. J.J.G. is an unpaid board member of the Rosetta Commons. Under institutional participation agreements between the University of Washington, acting on behalf of the Rosetta Commons, Johns Hopkins University may be entitled to a portion of revenue received on licensing Rosetta software including methods discussed/developed in this study. As a member of the Scientific Advisory Board, J.J.G. has a financial interest in Cyrus Biotechnology. Cyrus Biotechnology distributes the Rosetta software, which may include methods developed in this study. These arrangements have been reviewed and approved by the Johns Hopkins University in accordance with its conflict-of-interest policies. Publisher Copyright: {\textcopyright} 2021 The Authors",
year = "2022",
month = feb,
day = "11",
doi = "10.1016/j.patter.2021.100406",
language = "English (US)",
volume = "3",
journal = "Patterns",
issn = "2666-3899",
publisher = "Cell Press",
number = "2",
}