Cargando…
Predicting antibody binders and generating synthetic antibodies using deep learning
The antibody drug field has continually sought improvements to methods for candidate discovery and engineering. Historically, most such methods have been laboratory-based, but informatics methods have recently started to make an impact. Deep learning, a subfield of machine learning, is rapidly gaini...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Taylor & Francis
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9067455/ https://www.ncbi.nlm.nih.gov/pubmed/35482911 http://dx.doi.org/10.1080/19420862.2022.2069075 |
_version_ | 1784700009365635072 |
---|---|
author | Lim, Yoong Wearn Adler, Adam S. Johnson, David S. |
author_facet | Lim, Yoong Wearn Adler, Adam S. Johnson, David S. |
author_sort | Lim, Yoong Wearn |
collection | PubMed |
description | The antibody drug field has continually sought improvements to methods for candidate discovery and engineering. Historically, most such methods have been laboratory-based, but informatics methods have recently started to make an impact. Deep learning, a subfield of machine learning, is rapidly gaining prominence in the biomedical research. Recent advances in microfluidics technologies and next-generation sequencing have not only revolutionized therapeutic antibody discovery, but also contributed to a vast amount of antibody repertoire sequencing data, providing opportunities for deep learning-based applications. Previously, we used microfluidics, yeast display, and deep sequencing to generate a panel of binder and non-binder antibody sequences to the cancer immunotherapy targets PD-1 and CTLA-4. Here we encoded the antibody light and heavy chain complementarity-determining regions (CDR3s) into antibody images, then built and trained convolutional neural network models to classify binders and non-binders. To improve model interpretability, we performed in silico mutagenesis to identify CDR3 residues that were important for binder classification. We further built generative deep learning models using generative adversarial network models to produce synthetic antibodies against PD-1 and CTLA-4. Our models generated variable length CDR3 sequences that resemble real sequences. Overall, our study demonstrates that deep learning methods can be leveraged to mine and learn patterns in antibody sequences, offering insights into antibody engineering, optimization, and discovery. |
format | Online Article Text |
id | pubmed-9067455 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | Taylor & Francis |
record_format | MEDLINE/PubMed |
spelling | pubmed-90674552022-05-05 Predicting antibody binders and generating synthetic antibodies using deep learning Lim, Yoong Wearn Adler, Adam S. Johnson, David S. MAbs Report The antibody drug field has continually sought improvements to methods for candidate discovery and engineering. Historically, most such methods have been laboratory-based, but informatics methods have recently started to make an impact. Deep learning, a subfield of machine learning, is rapidly gaining prominence in the biomedical research. Recent advances in microfluidics technologies and next-generation sequencing have not only revolutionized therapeutic antibody discovery, but also contributed to a vast amount of antibody repertoire sequencing data, providing opportunities for deep learning-based applications. Previously, we used microfluidics, yeast display, and deep sequencing to generate a panel of binder and non-binder antibody sequences to the cancer immunotherapy targets PD-1 and CTLA-4. Here we encoded the antibody light and heavy chain complementarity-determining regions (CDR3s) into antibody images, then built and trained convolutional neural network models to classify binders and non-binders. To improve model interpretability, we performed in silico mutagenesis to identify CDR3 residues that were important for binder classification. We further built generative deep learning models using generative adversarial network models to produce synthetic antibodies against PD-1 and CTLA-4. Our models generated variable length CDR3 sequences that resemble real sequences. Overall, our study demonstrates that deep learning methods can be leveraged to mine and learn patterns in antibody sequences, offering insights into antibody engineering, optimization, and discovery. Taylor & Francis 2022-04-28 /pmc/articles/PMC9067455/ /pubmed/35482911 http://dx.doi.org/10.1080/19420862.2022.2069075 Text en © 2022 GigaGen, Inc. Published with license by Taylor & Francis Group, LLC. https://creativecommons.org/licenses/by-nc/4.0/This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/ (https://creativecommons.org/licenses/by-nc/4.0/) ), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. |
spellingShingle | Report Lim, Yoong Wearn Adler, Adam S. Johnson, David S. Predicting antibody binders and generating synthetic antibodies using deep learning |
title | Predicting antibody binders and generating synthetic antibodies using deep learning |
title_full | Predicting antibody binders and generating synthetic antibodies using deep learning |
title_fullStr | Predicting antibody binders and generating synthetic antibodies using deep learning |
title_full_unstemmed | Predicting antibody binders and generating synthetic antibodies using deep learning |
title_short | Predicting antibody binders and generating synthetic antibodies using deep learning |
title_sort | predicting antibody binders and generating synthetic antibodies using deep learning |
topic | Report |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9067455/ https://www.ncbi.nlm.nih.gov/pubmed/35482911 http://dx.doi.org/10.1080/19420862.2022.2069075 |
work_keys_str_mv | AT limyoongwearn predictingantibodybindersandgeneratingsyntheticantibodiesusingdeeplearning AT adleradams predictingantibodybindersandgeneratingsyntheticantibodiesusingdeeplearning AT johnsondavids predictingantibodybindersandgeneratingsyntheticantibodiesusingdeeplearning |