Computational design of antibodies

Fischman, Sharon; Ofran, Yishai

doi:10.1016/j.sbi.2018.04.007

Cited by 55 publications

(52 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, because of recent advances in computational power and algorithms, computational design is becoming an alternative method in antibody engineering. [13][14][15][16][17][18] One of the advantages to computational methods is that their use can be a rational approach when combined with a structure. Antibodies and their structures should be governed by physical laws, and based on physical principles, we should be able to predict behavior of antibodies in solution and in our body.…”

Section: Introductionmentioning

confidence: 99%

Engineering Stability, Viscosity, and Immunogenicity of Antibodies by Computational Design

Kuroda

Tsumoto

2020

Journal of Pharmaceutical Sciences

View full text Add to dashboard Cite

In recent years, computational methods have garnered much attention in protein engineering. A large number of computational methods have been developed to analyze the sequences and structures of proteins and have been used to predict the various properties. Antibodies are one of the emergent protein therapeutics, and thus, methods to control their physicochemical properties are highly desirable. However, despite the tremendous efforts of past decades, computational methods to predict the physicochemical properties of antibodies are still in their infancy. Experimental validations are certainly required for real-world applications, and the results should be interpreted with caution. Among the various properties of antibodies, we focus in this review on stability, viscosity, and immunogenicity, and we present the current status of computational methods to engineer such properties.

show abstract

Section: Introductionmentioning

confidence: 99%

Engineering Stability, Viscosity, and Immunogenicity of Antibodies by Computational Design

Kuroda

Tsumoto

2020

Journal of Pharmaceutical Sciences

View full text Add to dashboard Cite

show abstract

“…The need for computational modeling of protein binders is reflected by the constant interest in antibody docking and design [25][26][27]. Modeling antibodies, however, is particularly challenging because their binding mode involves the interaction of six complementarity-determining loops, some of which are considerably flexible.…”

mentioning

confidence: 99%

Computational Modeling of Designed Ankyrin Repeat Protein Complexes with Their Targets

Radom

Paci

Plückthun

2019

Journal of Molecular Biology

View full text Add to dashboard Cite

Recombinant therapeutic proteins are playing an ever-increasing role in the clinic. Highaffinity binding candidates can be produced in a high-throughput manner through the process of selection and evolution from large libraries, but the structures of the complexes with target protein can only be determined for a small number of them in a costly, low-throughput manner, typically by x-ray crystallography. Reliable modeling of complexes would greatly help to understand their mode of action, and improve them by further engineering, e.g., by designing bi-paratopic binders. Designed Ankyrin Repeat Proteins (DARPins) are one such class of antibody mimetics that have proven useful in the clinic, in diagnostics and research. Here we have developed a standardized procedure to model DARPin-target complexes that can be used to predict the structures of unknown complexes. It requires only the sequence of a DARPin and a structure of the unbound target. The procedure includes homology modeling of the DARPin, modeling of the flexible parts of a target, rigid body docking to ensembles of the target and docking with a partially flexible backbone. For a set of diverse DARPin-target complexes tested it generated a single model of the complex that well approximates the native state of the complex. We provide a protocol that can be used in a semi-automated way, and with tools that are freely available. The presented concepts should help to accelerate the development of novel bio-therapeutics for scaffolds with similar properties.

show abstract

“…Most computational methods for antibody design optimization assume that the structure of a target is known, however target structure is unknown is many cases 6 . Other approaches seek to optimize antibody properties primarily focused on predicting the structural conformation of the CDR-H3 loop, which to date remains a difficult unsolved challenge 7,8 .…”

Section: Main Textmentioning

confidence: 99%

Antibody Complementarity Determining Region Design Using High-Capacity Machine Learning

Liu

Zeng

Mueller

et al. 2019

Preprint

View full text Add to dashboard Cite

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. SignificanceAntibody 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.

show abstract

Computational design of antibodies

Cited by 55 publications

References 48 publications

Engineering Stability, Viscosity, and Immunogenicity of Antibodies by Computational Design

Engineering Stability, Viscosity, and Immunogenicity of Antibodies by Computational Design

Computational Modeling of Designed Ankyrin Repeat Protein Complexes with Their Targets

Antibody Complementarity Determining Region Design Using High-Capacity Machine Learning

Contact Info

Product

Resources

About