Antibody complementarity determining region design using high-capacity machine learning

By Ge Liu, Haoyang Zeng, Jonas W. Mueller, Brandon Carter, Ziheng Wang, J. Schilz, Geraldine Horny, M. Birnbaum, Stefan Ewert, David K. Gifford. 2019

172 citations recorded by the source corpus.

The precise targeting of antibodies and other protein therapeutics is required for their proper function and the elimination of deleterious off-target effects. Often the molecular structure of a therapeutic target is unknown and randomized methods are used to design antibodies without a model that relates antibody sequence to desired properties. Here we present a machine learning method that can design human Immunoglobulin G (IgG) antibodies with target affinities that are superior to candidates from phage display panning experiments within a limited design budget. We also demonstrate that machine learning can improve target-specificity by the modular composition of models from different experimental campaigns, enabling a new integrative approach to improving target specificity. Our results suggest a new path for the discovery of therapeutic molecules by demonstrating that predictive and differentiable models of antibody binding can be learned from high-throughput experimental data without the need for target structural data. Significance Antibody based therapeutics must meet both affinity and specificity metrics, and existing in vitro methods for meeting these metrics are based upon randomization and empirical testing. We demonstrate that with sufficient target-specific training data machine learning can suggest novel antibody variable domain sequences that are superior to those observed during training. Our machine learning method does not require any target structural information. We further show that data from disparate antibody campaigns can be combined by machine learning to improve antibody specificity.

Subfields: Molecular Biology, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Molecular Biology

Citations in the corpus

Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-design (2021, background): “Nevertheless, these physics-based methods are computationally expensive (Ingraham et al., 2019) and our desired objective can be much more complicated than low binding energy (Liu et al., 2020).”
A Review of Deep Learning Methods for Antibodies (2020, methodology): “Another sequence optimization algorithm, Ens-Grad, uses an ensemble of neural networks and gradient ascent to optimize an H3 seed sequence [36].”
Model-based reinforcement learning for biological sequence design (2020, methodology): “Recently, such an approach was used to optimize the binding affinity of IgG antibodies (Liu et al., 2019), where a neural network ensemble was used for f ′ w(x).”
Guided Generative Protein Design using Regularized Transformers (2022, methodology): “While other encoding methods that rely on convolutional and recurrent architectures have demonstrated success in this domain Liu et al. [2020], Alley et al. [2019], we choose to use a transformer for several key reasons.”