Bayesian Optimization of Antibodies Informed by a Generative Model of Evolving Sequences

• URL: http://arxiv.org/abs/2412.07763v1
• Date: Tue, 10 Dec 2024 18:57:48 GMT
• Title: Bayesian Optimization of Antibodies Informed by a Generative Model of Evolving Sequences
• Authors: Alan Nawzad Amin, Nate Gruver, Yilun Kuang, Lily Li, Hunter Elliott, Calvin McCarter, Aniruddh Raghu, Peyton Greenside, Andrew Gordon Wilson,
• Abstract summary: To build effective therapeutics, biologists iteratively mutate antibody sequences to improve binding and stability. Proposed mutations can be informed by previous measurements or by learning from large antibody databases to predict only typical antibodies. We introduce Clone-informed Bayesian Optimization (CloneBO), a Bayesian optimization procedure that efficiently optimize antibodies in the lab.
• Score: 36.58091564004384
• License:
• Abstract: To build effective therapeutics, biologists iteratively mutate antibody sequences to improve binding and stability. Proposed mutations can be informed by previous measurements or by learning from large antibody databases to predict only typical antibodies. Unfortunately, the space of typical antibodies is enormous to search, and experiments often fail to find suitable antibodies on a budget. We introduce Clone-informed Bayesian Optimization (CloneBO), a Bayesian optimization procedure that efficiently optimizes antibodies in the lab by teaching a generative model how our immune system optimizes antibodies. Our immune system makes antibodies by iteratively evolving specific portions of their sequences to bind their target strongly and stably, resulting in a set of related, evolving sequences known as a clonal family. We train a large language model, CloneLM, on hundreds of thousands of clonal families and use it to design sequences with mutations that are most likely to optimize an antibody within the human immune system. We propose to guide our designs to fit previous measurements with a twisted sequential Monte Carlo procedure. We show that CloneBO optimizes antibodies substantially more efficiently than previous methods in realistic in silico experiments and designs stronger and more stable binders in in vitro wet lab experiments.

Related papers

• Leveraging Large Language Models to Predict Antibody Biological Activity Against Influenza A Hemagglutinin [0.15547733154162566]
  We develop an AI model for predicting the binding and receptor blocking activity of antibodies against influenza A hemagglutininin (HA) antigens. Our models achieved an AUROC $geq$ 0.91 for predicting the activity of existing antibodies against seen HAs and an AUROC of 0.9 for unseen HAs.
  arXiv  Detail & Related papers  (2025-02-02T06:48:45Z)
• AlignAb: Pareto-Optimal Energy Alignment for Designing Nature-Like Antibodies [4.679586996508103]
  We present a three-stage framework for training deep learning models specializing in antibody sequence-structure co-design. We employ the learned representations to guide the training of a diffusion model for joint optimization over both sequence and structure of antibodies.
  arXiv  Detail & Related papers  (2024-12-30T14:50:32Z)
• Retrieval Augmented Diffusion Model for Structure-informed Antibody Design and Optimization [8.546688995090491]
  Antibodies are essential proteins responsible for immune responses in organisms. Recent advances in generative models have significantly enhanced rational antibody design. We propose a retrieval-augmented diffusion framework, termed RADAb, for efficient antibody design.
  arXiv  Detail & Related papers  (2024-10-19T08:53:01Z)
• Opponent Shaping for Antibody Development [49.26728828005039]
  Anti-viral therapies are typically designed to target only the current strains of a virus. therapy-induced selective pressures act on viruses to drive the emergence of mutated strains, against which initial therapies have reduced efficacy. We build on a computational model of binding between antibodies and viral antigens to implement a genetic simulation of viral evolutionary escape.
  arXiv  Detail & Related papers  (2024-09-16T14:56:27Z)
• Large scale paired antibody language models [40.401345152825314]
  We present IgBert and IgT5, the best performing antibody-specific language models developed to date. These models are trained comprehensively using the more than two billion Observed Space dataset. This advancement marks a significant leap forward in leveraging machine learning, large data sets and high-performance computing for enhancing antibody design for therapeutic development.
  arXiv  Detail & Related papers  (2024-03-26T17:21:54Z)
• Antigen-Specific Antibody Design via Direct Energy-based Preference Optimization [51.28231365213679]
  We tackle antigen-specific antibody sequence-structure co-design as an optimization problem towards specific preferences. We propose direct energy-based preference optimization to guide the generation of antibodies with both rational structures and considerable binding affinities to given antigens.
  arXiv  Detail & Related papers  (2024-03-25T09:41:49Z)
• AntiBARTy Diffusion for Property Guided Antibody Design [0.0]
  We train an antibody-specific language model, AntiBARTy, based on BART (Bidirectional and Auto-Regressive Transformer) and use its latent space to train a property-conditional diffusion model for guided IgG de novo design. As a test case, we show that we can effectively generate novel antibodies with improved in-silico solubility while maintaining antibody validity and controlling sequence diversity.
  arXiv  Detail & Related papers  (2023-09-22T18:30:50Z)
• xTrimoABFold: De novo Antibody Structure Prediction without MSA [77.47606749555686]
  We develop a novel model named xTrimoABFold to predict antibody structure from antibody sequence. The model was trained end-to-end on the antibody structures in PDB by minimizing the ensemble loss of domain-specific focal loss on CDR and the frame-aligned point loss.
  arXiv  Detail & Related papers  (2022-11-30T09:26:08Z)
• Incorporating Pre-training Paradigm for Antibody Sequence-Structure Co-design [134.65287929316673]
  Deep learning-based computational antibody design has attracted popular attention since it automatically mines the antibody patterns from data that could be complementary to human experiences. The computational methods heavily rely on high-quality antibody structure data, which is quite limited. Fortunately, there exists a large amount of sequence data of antibodies that can help model the CDR and alleviate the reliance on structure data.
  arXiv  Detail & Related papers  (2022-10-26T15:31:36Z)
• AntBO: Towards Real-World Automated Antibody Design with Combinatorial Bayesian Optimisation [53.43922443725598]
  We present AntBO: a Combinatorial optimisation algorithm enabling efficient in silico design of the CDRH3 region. To benchmark AntBO, we use the Absolut! software suite as a black-box oracle because it can score the target specificity and affinity of designed antibodies in silico. In under 200 protein designs, AntBO can suggest antibody sequences that outperform the best binding sequence drawn from 6.9 million experimentally obtained CDRH3s.
  arXiv  Detail & Related papers  (2022-01-29T12:03:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.