Animal Immunization, in Vitro Display Technologies, and Machine Learning for Antibody Discovery

Laustsen, Andreas Hougaard; Greiff, Victor; Karatt-Vellatt, Aneesh; Muyldermans, Serge; Jenkins, Timothy

doi:10.1016/j.tibtech.2021.03.003

Cited by 85 publications

(74 citation statements)

References 65 publications

(84 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For binary classification, the definition of non-binder (negative class / negative samples) as top 1%-5% versus 1%-100% high-affinity sequences had a significant impact on prediction accuracy (Figure 3E,F), suggesting that the former strategy better delineates the boundary between binders and non-binders. Similarly in the context of enrichment-based antibody screening (18,81), where each enrichment step defines "better binders", our results indicate that the non-binders in a late enrichment step could represent a good negative class in comparison with the non-binders in the early enrichment steps. Therefore, Absolut!…”

Section: Discussionsupporting

confidence: 55%

“…The generated database contains 1.1 billion antibody-antigen binding structures with conformational paratope, conformational epitope and affinity resolution (159 x 6.9x10 6 affinity matrix, Figure 1B). This dataset is several orders of magnitude larger than what is currently feasible to generate experimentally (18).…”

Section: Resultsmentioning

confidence: 99%

“…Machine learning (ML) represents an increasingly used approach for antibody-antigen binding prediction (3,10,18,(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32) given its capacity to infer the hidden nonlinear rules underlying high-complexity protein-protein interaction (30,(33)(34)(35)(36)(37)(38). Recent reports suggest that binding prediction based on antibody sequence or structure may be feasible as long as sufficiently large antigen(epitope)-specific antibody datasets become available (15,16,(39)(40)(41)(42)(43).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Unconstrained generation of synthetic antibody-antigen structures to guide machine learning methodology for real-world antibody specificity prediction

Akbar

Frank

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Machine learning (ML) is a key technology to enable accurate prediction of antibody-antigen binding, a prerequisite for in silico vaccine and antibody design. Two orthogonal problems hinder the current application of ML to antibody-specificity prediction and the benchmarking thereof: (i) The lack of a unified formalized mapping of immunological antibody specificity prediction problems into ML notation and (ii) the unavailability of large-scale training datasets. Here, we developed the Absolut! software suite that allows the parameter-based unconstrained generation of synthetic lattice-based 3D-antibody-antigen binding structures with ground-truth access to conformational paratope, epitope, and affinity. We show that Absolut!-generated datasets recapitulate critical biological sequence and structural features that render antibody-antigen binding prediction challenging. To demonstrate the immediate, high-throughput, and large-scale applicability of Absolut!, we have created an online database of 1 billion antibody-antigen structures, the extension of which is only constrained by moderate computational resources. We translated immunological antibody specificity prediction problems into ML tasks and used our database to investigate paratope-epitope binding prediction accuracy as a function of structural information encoding, dataset size, and ML method, which is unfeasible with existing experimental data. Furthermore, we found that in silico investigated conditions, predicted to increase antibody specificity prediction accuracy, align with and extend conclusions drawn from experimental antibody-antigen structural data. In summary, the Absolut! framework enables the development and benchmarking of ML strategies for biotherapeutics discovery and design.

show abstract

Section: Discussionsupporting

confidence: 55%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unconstrained generation of synthetic antibody-antigen structures to guide machine learning methodology for real-world antibody specificity prediction

Akbar

Frank

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therapeutic antibodies can be discovered via in vivo , in vitro or in silico approaches [ 1 , 2 ]. In vivo approach relies on the immunization of wild type or transgenic animals carrying human antibody gene segments, while in vitro approach employs the selection power by display technologies to pan large and diverse antibody libraries [ 3 ].…”

Section: Introductionmentioning

confidence: 99%

Antibody heavy chain CDR3 length-dependent usage of human IGHJ4 and IGHJ6 germline genes

Wang

Yan

Yang

et al. 2021

Antibody Therapeutics

View full text Add to dashboard Cite

Background Therapeutic antibody discovery using synthetic diversity has been proved productive, especially for target proteins not suitable for traditional animal immunization-based antibody discovery approaches. In recent years, many lines of evidences suggest that the quality of synthetic diversity design limits the development success of synthetic antibody hits. The aim of our study is to understand the quality limitation and to properly address the challenges with a better design. Methods Using VH3–23 as a model framework, we analyzed the naturally and productively rearranged CDR-H3 diversity in human immune repertoire. With homology modeling, we further built VH3–23-based structural models to understand the spatial paratope and its influencing parameters. Results We observed and quantitatively mapped CDR-H3 loop length-dependent usage of human IGHJ4 and IGHJ6 germline genes in the natural human immune repertoire. Skewed usage of DH2-JH6 and DH3-JH6 rearrangements was quantitatively determined in a CDR-H3 length-dependent manner in natural human antibodies with long CDR-H3 loops. Such CDR-H3 length dependent usage of human germline genes was not impacted by the choices of VH in the V(D)J recombination, ethnic background and health conditions. Structural modeling suggests choices of JH help to stabilize antibody CDR-H3 loop and JH only partially contributes to the paratope. Conclusions We quantitatively determined the CDR-H3 length-dependent usage of human germline genes, which makes it possible to design synthetic diversity fully mimicking that of natural immune repertoire. Our observations shed light on the design of next generation synthetic diversity with improved probability of success.

show abstract

“…And recently, immense efforts to utilize mAbs for the neutralization of viral agents, such as HIV, influenza, and SARS-CoV-2 (2-4) are ongoing as well. So far, however, lead times to mAb discovery and design are on average >3 years (5)(6)(7)(8). The reason for this is that current mAb development pipelines mostly rely on a combination of large screening libraries and experimental heuristics with very little to no emphasis on rule-driven discovery (9).…”

Section: Introductionmentioning

confidence: 99%

In silico proof of principle of machine learning-based antibody design at unconstrained scale

Akbar

Weber

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Generative machine learning (ML) has been postulated to be a major driver in the computational design of antigen-specific monoclonal antibodies (mAb). However, efforts to confirm this hypothesis have been hindered by the infeasibility of testing arbitrarily large numbers of antibody sequences for their most critical design parameters: paratope, epitope, affinity, and developability. To address this challenge, we leveraged a lattice-based antibody-antigen binding simulation framework, which incorporates a wide range of physiological antibody binding parameters. The simulation framework enables both the computation of antibody-antigen 3D-structures as well as functions as an oracle for unrestricted prospective evaluation of the antigen specificity of ML-generated antibody sequences. We found that a deep generative model, trained exclusively on antibody sequence (1D) data can be used to design native-like conformational (3D) epitope-specific antibodies, matching or exceeding the training dataset in affinity and developability variety. Furthermore, we show that transfer learning enables the generation of high-affinity antibody sequences from low-N training data. Finally, we validated that the antibody design insight gained from simulated antibody-antigen binding data is applicable to experimental real-world data. Our work establishes a priori feasibility and the theoretical foundation of high-throughput ML-based mAb design.HighlightsA large-scale dataset of 70M [3 orders of magnitude larger than the current state of the art] synthetic antibody-antigen complexes, that reflect biological complexity, allows the prospective evaluation of antibody generative deep learningCombination of generative learning, synthetic antibody-antigen binding data, and prospective evaluation shows that deep learning driven antibody design and discovery at an unconstrained level is feasibleTransfer learning (low-N learning) coupled to generative learning shows that antibody-binding rules may be transferred across unrelated antibody-antigen complexesExperimental validation of antibody-design conclusions drawn from deep learning on synthetic antibody-antigen binding dataGraphical abstractWe leverage large synthetic ground-truth data to demonstrate the (A,B) unconstrained deep generative learning-based generation of native-like antibody sequences, (C) the prospective evaluation of conformational (3D) affinity, paratope-epitope pairs, and developability. (D) Finally, we show increased generation quality of low-N-based machine learning models via transfer learning.

show abstract

Animal Immunization, in Vitro Display Technologies, and Machine Learning for Antibody Discovery

Cited by 85 publications

References 65 publications

Unconstrained generation of synthetic antibody-antigen structures to guide machine learning methodology for real-world antibody specificity prediction

Unconstrained generation of synthetic antibody-antigen structures to guide machine learning methodology for real-world antibody specificity prediction

Antibody heavy chain CDR3 length-dependent usage of human IGHJ4 and IGHJ6 germline genes

In silico proof of principle of machine learning-based antibody design at unconstrained scale

Contact Info

Product

Resources

About