A dataset comprised of binding interactions for 104,972 antibodies against a SARS-CoV-2 peptide

Engelhart, Emily; Emerson, Ryan; Shing, Leslie; Lennartz, Chelsea; Guion, Daniel; Kelley, Mary; Lin, Charles; Lopez, Randolph; Younger, David; Walsh, Matthew E.

doi:10.1038/s41597-022-01779-4

Cited by 20 publications

(25 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The antibody covid specificity Dataset (AlphaSeq) [40] was downloaded from https://zenodo.org/record/5095284. AlphaSeq consists of 71,415 paired heavy and light chains, with continuous measurements of their affinity to an epitope of the SARS-CoV-2 spike protein.…”

Section: Methodsmentioning

confidence: 99%

“…We evaluate PLM-based antibody affinity predictions through the lens of the AlphaSeq dataset [40], which provides a continuous label for the estimated binding affinity of antibodies to a specific SARS-CoV-2 epitope, including several replicate measurements for many sequences. We use PLMs to compute fixed-size embeddings of the antibody protein sequences, which reduces the task of predicting the affinity to a supervised regression problem, mapping embedding vectors to a continuous number.…”

Section: Model Architecture For Antibody Specificity Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Do Domain-Specific Protein Language Models Outperform General Models on Immunology-Related Tasks?

Deutschmann,

Pelissier,

Weber

et al. 2023

Preprint

View full text Add to dashboard Cite

Deciphering the antigen recognition capabilities by T~cell and B~cell receptors (antibodies) is essential for advancing our understanding of adaptive immune system responses. In recent years, the development of protein language models (PLMs) has facilitated the development of bioinformatic pipelines where complex amino acid sequences are transformed into vectorized embeddings, which are then applied to a range of downstream analytical tasks. With their success, we have witnessed the emergence of domain-specific PLMs tailored to specific proteins, such as immune receptors. Domain-specific models are often assumed to possess enhanced representation capabilities for targeted applications, however, this assumption has not been thoroughly evaluated. In this manuscript, we assess the efficacy of both generalist and domain-specific transformer-based embeddings in characterizing B and T cell receptors. Specifically, we assess the accuracy of models that leverage these embeddings to predict antigen specificity and elucidate the evolutionary changes that B cells undergo during an immune response. We demonstrate that the prevailing notion of domain-specific models outperforming general models requires a more nuanced examination. We also observe remarkable differences between generalist and domain-specific PLMs, not only in terms of performance but also in the manner they encode information. Finally, we observe that the choice of the size and the embedding layer in PLMs are essential model hyperparameters in different tasks. Overall, our analyzes reveal the promising potential of PLMs in modeling protein function while providing insights into their information handling capabilities. We also discuss the crucial factors that should be taken into account when selecting a PLM tailored to a particular task.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Model Architecture For Antibody Specificity Predictionmentioning

confidence: 99%

Do Domain-Specific Protein Language Models Outperform General Models on Immunology-Related Tasks?

Deutschmann,

Pelissier,

Weber

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We generated our supervised training data using an engineered yeast mating assay and have published it separately to support its reuse [23]. This data includes 17,118 heavy chain scFv binding measurements and 26,223 light chain scFv binding measurements against a single target peptide and were collected in pooled, yeast-based mating assays.…”

Section: Resultsmentioning

confidence: 99%

“…Experimental Binding Measurements for Sequence-to-Affinity Model Training. We separately published the training data to support ease-of-reuse [23]. Briefly, experimental measurements were made, in technical triplicate, by A-Alpha Bio, LLC and are a "predicted affinity" value [27].…”

Section: Methodsmentioning

confidence: 99%

Machine Learning Optimization of Candidate Antibodies Yields Highly Diverse Sub-nanomolar Affinity Antibody Libraries

Gupta

Spaeth

et al. 2022

Preprint

View full text Add to dashboard Cite

Therapeutic antibodies are an important and rapidly growing drug modality. However, the design and discovery of early-stage antibody therapeutics remain a time and cost-intensive endeavor. In this work, we present an end-to-end Bayesian, language model-based method for designing large and diverse libraries of high-affinity single-chain variable fragments (scFvs). We integrate target-specific binding affinities with information from millions of natural protein sequences in a probabilistic machine learning framework to design thousands of scFvs that are then empirically measured. In a head-to-head comparison with a directed evolution approach, we show that the best scFv generated from our method represents a 28.8-fold improvement in binding over the best scFv from the directed evolution. Additionally, 99% of the designed scFvs in our most successful library are improvements over the initial candidate scFv. By comparing a library's predicted success to actual measurements, we demonstrate our method's ability to explore tradeoffs between library success and diversity during the design phase and prior to experimental testing. The results of our work highlight the significant impact machine learning models can have on scFv development. We expect our end-to-end method to be broadly applicable and able to provide value to other protein engineering tasks.

show abstract

“…14H and 14L: The 14H and 14L datasets are sourced from the LL-SARS-CoV-2 database 36 . This database was designed to choose two heavy chains and two light chains from three main chains of antibodies as the framework for constructing an antibody library.…”

Section: Methodsmentioning

confidence: 99%

De novogeneration of antibody CDRH3 with a pre-trained generative large language model

He,

Guan

et al. 2023

Preprint

View full text Add to dashboard Cite

Artificial Intelligence (AI) techniques have made great advances in assisting antibody design. However, antibody design still heavily relies on isolating antigen-specific antibodies from serum, which is a resource-intensive and time-consuming process. To address this issue, we propose a Pre-trained Antibody generative large Language Model (PALM) for the de novo generation of artificial antibodies heavy chain complementarity-determining region 3 (CDRH3) with desired antigen-binding specificity, reducing the reliance on natural antibodies. We also build a high-precision model antigen-antibody binder (A2binder) that pairs antigen epitope sequences with antibody sequences to predict binding specificity and affinity. PALM-generated antibodies exhibit binding ability to SARS-CoV-2 antigens, including the emerging XBB variant, as confirmed through in-silico analysis and in-vitro assays. The in-vitro assays validated that PALM-generated antibodies achieve high binding affinity and potent neutralization capability against both wild-type and XBB spike proteins of SARS-CoV-2. Meanwhile, A2binder demonstrated exceptional predictive performance on binding specificity for various epitopes and variants. Furthermore, by incorporating the attention mechanism into the PALM model, we have improved its interpretability, providing crucial insights into the fundamental principles of antibody design.

show abstract

A dataset comprised of binding interactions for 104,972 antibodies against a SARS-CoV-2 peptide

Cited by 20 publications

References 26 publications

Do Domain-Specific Protein Language Models Outperform General Models on Immunology-Related Tasks?

Do Domain-Specific Protein Language Models Outperform General Models on Immunology-Related Tasks?

Machine Learning Optimization of Candidate Antibodies Yields Highly Diverse Sub-nanomolar Affinity Antibody Libraries

De novogeneration of antibody CDRH3 with a pre-trained generative large language model

Contact Info

Product

Resources

About