Igor Snapkov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Igor Snapkov has a total of 158 co-authors. Here are the top co-authors by number of shared publications:
- Victor Greiff
University of Oslo
15 shared publications - Geir Kjetil Sandve
Department of Informatics, University of Oslo, Norway
13 shared publications - Rahmad Akbar
Department of Immunology, University of Oslo, Oslo, Oslo, 0372
12 shared publications - Milena Pavlović
Department of Informatics, University of Oslo, Norway
11 shared publications - Cédric R. Weber
Department of Biosystems Science and Engineering, ETH Zurich, Basel, Switzerland
10 shared publications - Maria Chernigovskaya
Department of Immunology, University of Oslo
9 shared publications
Latest Publications
- Open Access
- 26 March 2025
- Published by Springer Nature in Archives of Toxicology
- Vol. 99 (5) , 1791-1841
Hazard characterization of the mycotoxins enniatins and beauvericin to identify data gaps and improve risk assessment for human health
Enniatins (ENNs) and beauvericin (BEA) are cyclic hexadepsipeptide fungal metabolites which have demonstrated antibiotic, antimycotic, and insecticidal activities. The substantial toxic potentials of these mycotoxins are associated with their ionophoric molecular properties and relatively high lipophilicities. ENNs occur extensively in grain and grain-derived products and are considered a food safety issue by the European Food Safety Authority (EFSA). The tolerable daily intake and maximum levels for ENNs in humans and animals remain unestablished due to key toxicological and toxicokinetic data gaps, preventing full risk assessment. Aiming to find critical data gaps impeding hazard characterization and risk evaluation, this review presents a comprehensive summary of the existing information from in vitro and in vivo studies on toxicokinetic characteristics and cytotoxic, genotoxic, immunotoxic, endocrine, reproductive and developmental effects of the most prevalent ENN analogues (ENN A, A1, B, B1) and BEA. The missing information identified showed that additional studies on ENNs and BEA have to be performed before sufficient data for an in-depth hazard characterisation of these mycotoxins become available.
- 26 March 2025
- Published by Springer Nature in Archives of Toxicology
- Vol. 99 (5) , 1791-1841
Open AccessHazard characterization of the mycotoxins enniatins and beauvericin to identify data gaps and improve risk assessment for human health
Enniatins (ENNs) and beauvericin (BEA) are cyclic hexadepsipeptide fungal metabolites which have demonstrated antibiotic, antimycotic, and insecticidal activities. The substantial toxic potentials of these mycotoxins are associated with their ionophoric molecular properties and relatively high lipophilicities. ENNs occur extensively in grain and grain-derived products and are considered a food safety issue by the European Food Safety Authority (EFSA). The tolerable daily intake and maximum levels for ENNs in humans and animals remain unestablished due to key toxicological and toxicokinetic data gaps, preventing full risk assessment. Aiming to find critical data gaps impeding hazard characterization and risk evaluation, this review presents a comprehensive summary of the existing information from in vitro and in vivo studies on toxicokinetic characteristics and cytotoxic, genotoxic, immunotoxic, endocrine, reproductive and developmental effects of the most prevalent ENN analogues (ENN A, A1, B, B1) and BEA. The missing information identified showed that additional studies on ENNs and BEA have to be performed before sufficient data for an in-depth hazard characterisation of these mycotoxins become available.
- Open Access
Systematic benchmarking of mass spectrometry-based antibody sequencing reveals methodological biases
The circulating antibody repertoire is crucial for immune protection, holding significant immunological and biotechnological value. While bottom-up mass spectrometry (MS) is the most widely used proteomics technique for profiling the sequence diversity of circulating antibodies (Ab-seq), it has not been thoroughly benchmarked. We quantified the replicability and robustness of Ab-seq using six monoclonal antibodies with known protein sequences in 70 different combinations of concentration and oligoclonality, both with and without polyclonal serum IgG background. Each combination underwent four protease treatments and was analyzed across four experimental and three technical replicates, totaling 3,360 LC-MS/MS runs. We quantified the dependence of MS-based Ab-seq identification on antibody sequence, concentration, protease, background signal diversity, and bioinformatics setups. Integrating the data from experimental replicates, proteases, and bioinformatics tools enhanced antibody identification. De novo peptide sequencing showed similar performance to database-dependent methods for higher antibody concentrations, but de novo antibody reconstruction remains challenging. Our work provides a foundational resource for the field of MS-based antibody profiling.
Open AccessSystematic benchmarking of mass spectrometry-based antibody sequencing reveals methodological biases
The circulating antibody repertoire is crucial for immune protection, holding significant immunological and biotechnological value. While bottom-up mass spectrometry (MS) is the most widely used proteomics technique for profiling the sequence diversity of circulating antibodies (Ab-seq), it has not been thoroughly benchmarked. We quantified the replicability and robustness of Ab-seq using six monoclonal antibodies with known protein sequences in 70 different combinations of concentration and oligoclonality, both with and without polyclonal serum IgG background. Each combination underwent four protease treatments and was analyzed across four experimental and three technical replicates, totaling 3,360 LC-MS/MS runs. We quantified the dependence of MS-based Ab-seq identification on antibody sequence, concentration, protease, background signal diversity, and bioinformatics setups. Integrating the data from experimental replicates, proteases, and bioinformatics tools enhanced antibody identification. De novo peptide sequencing showed similar performance to database-dependent methods for higher antibody concentrations, but de novo antibody reconstruction remains challenging. Our work provides a foundational resource for the field of MS-based antibody profiling.
- Open Access
- 30 July 2024
- Published by Springer Nature in Archives of Toxicology
- Vol. 98 (10) , 3513-3514
Author Correction: Hazard characterization of Alternaria toxins to identify data gaps and improve risk assessment for human health
- 30 July 2024
- Published by Springer Nature in Archives of Toxicology
- Vol. 98 (10) , 3513-3514
Open AccessAuthor Correction: Hazard characterization of Alternaria toxins to identify data gaps and improve risk assessment for human health
- Open Access
- 26 December 2023
- Published by Springer Nature in Archives of Toxicology
- Vol. 98 (2) , 425-469
Hazard characterization of Alternaria toxins to identify data gaps and improve risk assessment for human health
Fungi of the genus Alternaria are ubiquitous plant pathogens and saprophytes which are able to grow under varying temperature and moisture conditions as well as on a large range of substrates. A spectrum of structurally diverse secondary metabolites with toxic potential has been identified, but occurrence and relative proportion of the different metabolites in complex mixtures depend on strain, substrate, and growth conditions. This review compiles the available knowledge on hazard identification and characterization of Alternaria toxins. Alternariol (AOH), its monomethylether AME and the perylene quinones altertoxin I (ATX-I), ATX-II, ATX-III, alterperylenol (ALP), and stemphyltoxin III (STTX-III) showed in vitro genotoxic and mutagenic properties. Of all identified Alternaria toxins, the epoxide-bearing analogs ATX-II, ATX-III, and STTX-III show the highest cytotoxic, genotoxic, and mutagenic potential in vitro. Under hormone-sensitive conditions, AOH and AME act as moderate xenoestrogens, but in silico modeling predicts further Alternaria toxins as potential estrogenic factors. Recent studies indicate also an immunosuppressive role of AOH and ATX-II; however, no data are available for the majority of Alternaria toxins. Overall, hazard characterization of Alternaria toxins focused, so far, primarily on the commercially available dibenzo-α-pyrones AOH and AME and tenuazonic acid (TeA). Limited data sets are available for altersetin (ALS), altenuene (ALT), and tentoxin (TEN). The occurrence and toxicological relevance of perylene quinone-based Alternaria toxins still remain to be fully elucidated. We identified data gaps on hazard identification and characterization crucial to improve risk assessment of Alternaria mycotoxins for consumers and occupationally exposed workers.
- 26 December 2023
- Published by Springer Nature in Archives of Toxicology
- Vol. 98 (2) , 425-469
Open AccessHazard characterization of Alternaria toxins to identify data gaps and improve risk assessment for human health
Fungi of the genus Alternaria are ubiquitous plant pathogens and saprophytes which are able to grow under varying temperature and moisture conditions as well as on a large range of substrates. A spectrum of structurally diverse secondary metabolites with toxic potential has been identified, but occurrence and relative proportion of the different metabolites in complex mixtures depend on strain, substrate, and growth conditions. This review compiles the available knowledge on hazard identification and characterization of Alternaria toxins. Alternariol (AOH), its monomethylether AME and the perylene quinones altertoxin I (ATX-I), ATX-II, ATX-III, alterperylenol (ALP), and stemphyltoxin III (STTX-III) showed in vitro genotoxic and mutagenic properties. Of all identified Alternaria toxins, the epoxide-bearing analogs ATX-II, ATX-III, and STTX-III show the highest cytotoxic, genotoxic, and mutagenic potential in vitro. Under hormone-sensitive conditions, AOH and AME act as moderate xenoestrogens, but in silico modeling predicts further Alternaria toxins as potential estrogenic factors. Recent studies indicate also an immunosuppressive role of AOH and ATX-II; however, no data are available for the majority of Alternaria toxins. Overall, hazard characterization of Alternaria toxins focused, so far, primarily on the commercially available dibenzo-α-pyrones AOH and AME and tenuazonic acid (TeA). Limited data sets are available for altersetin (ALS), altenuene (ALT), and tentoxin (TEN). The occurrence and toxicological relevance of perylene quinone-based Alternaria toxins still remain to be fully elucidated. We identified data gaps on hazard identification and characterization crucial to improve risk assessment of Alternaria mycotoxins for consumers and occupationally exposed workers.
- Open Access
Benchmarking and integrating human B-cell receptor genomic and antibody proteomic profiling
Immunoglobulins (Ig), which exist either as B-cell receptors (BCR) on the surface of B cells or as antibodies when secreted, play a key role in the recognition and response to antigenic threats. The capability to jointly characterize the BCR and antibody repertoire is crucial in understanding human adaptive immunity. From peripheral blood, bulk BCR sequencing (bulkBCR-seq) currently provides the highest sampling depth, single-cell BCR sequencing (scBCR-seq) allows for paired chain characterization, and antibody peptide sequencing by tandem mass spectrometry (Ab-seq) provides information on the composition of secreted antibodies in the serum. Although still rare, studies combining these three technologies would comprehensively capture the humoral immune response. Yet, it has not been benchmarked to what extent the datasets generated by these three technologies overlap and complement each other. To address this question, we isolated peripheral blood B cells from healthy donors and sequenced BCRs at bulk and single-cell level, in addition to utilizing publicly available sequencing data. Integrated analysis was performed on these datasets, resolved by replicates and across individuals. Simultaneously, serum antibodies were isolated, digested with multiple proteases, and analyzed with Ab-seq. Systems immunology analysis showed high concordance in repertoire features between bulk and scBCR-seq within individuals, especially when replicates were utilized. In addition, Ab-seq identified clonotype-specific peptides using both bulk and scBCR-seq library references, demonstrating the feasibility of combining scBCR-seq and Ab-seq for reconstructing paired-chain Ig sequences from the serum antibody repertoire. Collectively, our work serves as a proof-of-principle for combining bulk sequencing, single-cell sequencing, and mass spectrometry as complementary methods towards capturing humoral immunity in its entirety.
Open AccessBenchmarking and integrating human B-cell receptor genomic and antibody proteomic profiling
Immunoglobulins (Ig), which exist either as B-cell receptors (BCR) on the surface of B cells or as antibodies when secreted, play a key role in the recognition and response to antigenic threats. The capability to jointly characterize the BCR and antibody repertoire is crucial in understanding human adaptive immunity. From peripheral blood, bulk BCR sequencing (bulkBCR-seq) currently provides the highest sampling depth, single-cell BCR sequencing (scBCR-seq) allows for paired chain characterization, and antibody peptide sequencing by tandem mass spectrometry (Ab-seq) provides information on the composition of secreted antibodies in the serum. Although still rare, studies combining these three technologies would comprehensively capture the humoral immune response. Yet, it has not been benchmarked to what extent the datasets generated by these three technologies overlap and complement each other. To address this question, we isolated peripheral blood B cells from healthy donors and sequenced BCRs at bulk and single-cell level, in addition to utilizing publicly available sequencing data. Integrated analysis was performed on these datasets, resolved by replicates and across individuals. Simultaneously, serum antibodies were isolated, digested with multiple proteases, and analyzed with Ab-seq. Systems immunology analysis showed high concordance in repertoire features between bulk and scBCR-seq within individuals, especially when replicates were utilized. In addition, Ab-seq identified clonotype-specific peptides using both bulk and scBCR-seq library references, demonstrating the feasibility of combining scBCR-seq and Ab-seq for reconstructing paired-chain Ig sequences from the serum antibody repertoire. Collectively, our work serves as a proof-of-principle for combining bulk sequencing, single-cell sequencing, and mass spectrometry as complementary methods towards capturing humoral immunity in its entirety.
- Open Access
Biophysical cartography of the native and human-engineered antibody landscapes quantifies the plasticity of antibody developability
Designing effective monoclonal antibody (mAb) therapeutics faces a multi-parameter optimization challenge known as “developability”, which reflects an antibody’s ability to progress through development stages based on its physicochemical properties. While natural antibodies may provide valuable guidance for mAb selection, we lack a comprehensive understanding of natural developability parameter (DP) plasticity (redundancy, predictability, sensitivity) and how the DP landscapes of human-engineered and natural antibodies relate to one another. These gaps hinder fundamental developability profile cartography. To chart natural and engineered DP landscapes, we computed 40 sequence- and 46 structure-based DPs of over two million native and human-engineered single-chain antibody sequences. We found lower redundancy among structure-based compared to sequence-based DPs. Sequence DP sensitivity to single amino acid substitutions varied by antibody region and DP, and structure DP values varied across the conformational ensemble of antibody structures. Sequence DPs were more predictable than structure-based ones across different machine-learning tasks and embeddings, indicating a constrained sequence-based design space. Human-engineered antibodies were localized within the developability and sequence landscapes of natural antibodies, suggesting that human-engineered antibodies explore mere subspaces of the natural one. Our work quantifies the plasticity of antibody developability, providing a fundamental resource for multi-parameter therapeutic mAb design.
Open AccessBiophysical cartography of the native and human-engineered antibody landscapes quantifies the plasticity of antibody developability
Designing effective monoclonal antibody (mAb) therapeutics faces a multi-parameter optimization challenge known as “developability”, which reflects an antibody’s ability to progress through development stages based on its physicochemical properties. While natural antibodies may provide valuable guidance for mAb selection, we lack a comprehensive understanding of natural developability parameter (DP) plasticity (redundancy, predictability, sensitivity) and how the DP landscapes of human-engineered and natural antibodies relate to one another. These gaps hinder fundamental developability profile cartography. To chart natural and engineered DP landscapes, we computed 40 sequence- and 46 structure-based DPs of over two million native and human-engineered single-chain antibody sequences. We found lower redundancy among structure-based compared to sequence-based DPs. Sequence DP sensitivity to single amino acid substitutions varied by antibody region and DP, and structure DP values varied across the conformational ensemble of antibody structures. Sequence DPs were more predictable than structure-based ones across different machine-learning tasks and embeddings, indicating a constrained sequence-based design space. Human-engineered antibodies were localized within the developability and sequence landscapes of natural antibodies, suggesting that human-engineered antibodies explore mere subspaces of the natural one. Our work quantifies the plasticity of antibody developability, providing a fundamental resource for multi-parameter therapeutic mAb design.
- Open Access
- 4 April 2022
- Published by Taylor & Francis in mAbs
- Vol. 14 (1) , 2031482
In silico proof of principle of machine learning-based antibody design at unconstrained scale
Generative machine learning (ML) has been postulated to become 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 the computation of synthetic antibody-antigen 3D-structures, and it functions as an oracle for unrestricted prospective evaluation and benchmarking of antibody design parameters of ML-generated antibody sequences. We found that a deep generative model, trained exclusively on antibody sequence (one dimensional: 1D) data can be used to design conformational (three dimensional: 3D) epitope-specific antibodies, matching, or exceeding the training dataset in affinity and developability parameter value variety. Furthermore, we established a lower threshold of sequence diversity necessary for high-accuracy generative antibody ML and demonstrated that this lower threshold also holds on experimental real-world data. Finally, we show that transfer learning enables the generation of high-affinity antibody sequences from low-N training data. Our work establishes a priori feasibility and the theoretical foundation of high-throughput ML-based mAb design.
- 4 April 2022
- Published by Taylor & Francis in mAbs
- Vol. 14 (1) , 2031482
Open AccessIn silico proof of principle of machine learning-based antibody design at unconstrained scale
Generative machine learning (ML) has been postulated to become 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 the computation of synthetic antibody-antigen 3D-structures, and it functions as an oracle for unrestricted prospective evaluation and benchmarking of antibody design parameters of ML-generated antibody sequences. We found that a deep generative model, trained exclusively on antibody sequence (one dimensional: 1D) data can be used to design conformational (three dimensional: 3D) epitope-specific antibodies, matching, or exceeding the training dataset in affinity and developability parameter value variety. Furthermore, we established a lower threshold of sequence diversity necessary for high-accuracy generative antibody ML and demonstrated that this lower threshold also holds on experimental real-world data. Finally, we show that transfer learning enables the generation of high-affinity antibody sequences from low-N training data. Our work establishes a priori feasibility and the theoretical foundation of high-throughput ML-based mAb design.
- Open Access
Reference-based comparison of adaptive immune receptor repertoires
B- and T-cell receptor (immune) repertoires can represent an individual’s immune history. While current repertoire analysis methods aim to discriminate between health and disease states, they are typically based on only a limited number of parameters (e.g., clonal diversity, germline usage). Here, we introduce immuneREF: a quantitative multi-dimensional measure of adaptive immune repertoire (and transcriptome) similarity that allows interpretation of immune repertoire variation by relying on both repertoire features and cross-referencing of simulated and experimental datasets. immuneREF is implemented in an R package and was validated based on detection sensitivity of immune repertoires with known similarities and dissimilarities. To quantify immune repertoire similarity landscapes across health and disease, we applied immuneREF to >2400 datasets from individuals with varying immune states (healthy, [autoimmune] disease and infection [Covid-19], immune cell population). Importantly we discovered, in contrast to the current paradigm, that blood-derived immune repertoires of healthy and diseased individuals are highly similar for certain immune states, suggesting that repertoire changes to immune perturbations are less pronounced than previously thought. In conclusion, immuneREF implements population-wide analysis of immune repertoire similarity and thus enables the study of the adaptive immune response across health and disease states.
Open AccessReference-based comparison of adaptive immune receptor repertoires
B- and T-cell receptor (immune) repertoires can represent an individual’s immune history. While current repertoire analysis methods aim to discriminate between health and disease states, they are typically based on only a limited number of parameters (e.g., clonal diversity, germline usage). Here, we introduce immuneREF: a quantitative multi-dimensional measure of adaptive immune repertoire (and transcriptome) similarity that allows interpretation of immune repertoire variation by relying on both repertoire features and cross-referencing of simulated and experimental datasets. immuneREF is implemented in an R package and was validated based on detection sensitivity of immune repertoires with known similarities and dissimilarities. To quantify immune repertoire similarity landscapes across health and disease, we applied immuneREF to >2400 datasets from individuals with varying immune states (healthy, [autoimmune] disease and infection [Covid-19], immune cell population). Importantly we discovered, in contrast to the current paradigm, that blood-derived immune repertoires of healthy and diseased individuals are highly similar for certain immune states, suggesting that repertoire changes to immune perturbations are less pronounced than previously thought. In conclusion, immuneREF implements population-wide analysis of immune repertoire similarity and thus enables the study of the adaptive immune response across health and disease states.
- Open Access
- 23 November 2021
- Published by Cold Spring Harbor Laboratory in Genome Research
- Vol. 31 (12) , 2209-2224
Individualized VDJ recombination predisposes the available Ig sequence space
The process of recombination between variable (V), diversity (D), and joining (J) immunoglobulin (Ig) gene segments determines an individual's naive Ig repertoire and, consequently, (auto)antigen recognition. VDJ recombination follows probabilistic rules that can be modeled statistically. So far, it remains unknown whether VDJ recombination rules differ between individuals. If these rules differed, identical (auto)antigen-specific Ig sequences would be generated with individual-specific probabilities, signifying that the available Ig sequence space is individual specific. We devised a sensitivity-tested distance measure that enables inter-individual comparison of VDJ recombination models. We discovered, accounting for several sources of noise as well as allelic variation in Ig sequencing data, that not only unrelated individuals but also human monozygotic twins and even inbred mice possess statistically distinguishable immunoglobulin recombination models. This suggests that, in addition to genetic, there is also nongenetic modulation of VDJ recombination. We demonstrate that population-wide individualized VDJ recombination can result in orders of magnitude of difference in the probability to generate (auto)antigen-specific Ig sequences. Our findings have implications for immune receptor–based individualized medicine approaches relevant to vaccination, infection, and autoimmunity.
- 23 November 2021
- Published by Cold Spring Harbor Laboratory in Genome Research
- Vol. 31 (12) , 2209-2224
Open AccessIndividualized VDJ recombination predisposes the available Ig sequence space
The process of recombination between variable (V), diversity (D), and joining (J) immunoglobulin (Ig) gene segments determines an individual's naive Ig repertoire and, consequently, (auto)antigen recognition. VDJ recombination follows probabilistic rules that can be modeled statistically. So far, it remains unknown whether VDJ recombination rules differ between individuals. If these rules differed, identical (auto)antigen-specific Ig sequences would be generated with individual-specific probabilities, signifying that the available Ig sequence space is individual specific. We devised a sensitivity-tested distance measure that enables inter-individual comparison of VDJ recombination models. We discovered, accounting for several sources of noise as well as allelic variation in Ig sequencing data, that not only unrelated individuals but also human monozygotic twins and even inbred mice possess statistically distinguishable immunoglobulin recombination models. This suggests that, in addition to genetic, there is also nongenetic modulation of VDJ recombination. We demonstrate that population-wide individualized VDJ recombination can result in orders of magnitude of difference in the probability to generate (auto)antigen-specific Ig sequences. Our findings have implications for immune receptor–based individualized medicine approaches relevant to vaccination, infection, and autoimmunity.
- Open Access
In silico proof of principle of machine learning-based antibody design at unconstrained scale
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. Highlights A 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 learning Combination 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 feasible Transfer learning (low-N learning) coupled to generative learning shows that antibody-binding rules may be transferred across unrelated antibody-antigen complexes Experimental validation of antibody-design conclusions drawn from deep learning on synthetic antibody-antigen binding data We 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.
Open AccessIn silico proof of principle of machine learning-based antibody design at unconstrained scale
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. Highlights A 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 learning Combination 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 feasible Transfer learning (low-N learning) coupled to generative learning shows that antibody-binding rules may be transferred across unrelated antibody-antigen complexes Experimental validation of antibody-design conclusions drawn from deep learning on synthetic antibody-antigen binding data We 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.