Antibody design using LSTM based deep generative model from phage display library for affinity maturation.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
12 03 2021
12 03 2021
Historique:
received:
07
08
2020
accepted:
26
02
2021
entrez:
13
3
2021
pubmed:
14
3
2021
medline:
15
12
2021
Statut:
epublish
Résumé
Molecular evolution is an important step in the development of therapeutic antibodies. However, the current method of affinity maturation is overly costly and labor-intensive because of the repetitive mutation experiments needed to adequately explore sequence space. Here, we employed a long short term memory network (LSTM)-a widely used deep generative model-based sequence generation and prioritization procedure to efficiently discover antibody sequences with higher affinity. We applied our method to the affinity maturation of antibodies against kynurenine, which is a metabolite related to the niacin synthesis pathway. Kynurenine binding sequences were enriched through phage display panning using a kynurenine-binding oriented human synthetic Fab library. We defined binding antibodies using a sequence repertoire from the NGS data to train the LSTM model. We confirmed that likelihood of generated sequences from a trained LSTM correlated well with binding affinity. The affinity of generated sequences are over 1800-fold higher than that of the parental clone. Moreover, compared to frequency based screening using the same dataset, our machine learning approach generated sequences with greater affinity.
Identifiants
pubmed: 33712669
doi: 10.1038/s41598-021-85274-7
pii: 10.1038/s41598-021-85274-7
pmc: PMC7955064
doi:
Substances chimiques
Antibodies
0
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
5852Références
Protein Expr Purif. 2016 Oct;126:1-8
pubmed: 27181246
Nucleic Acids Res. 2015 Jan;43(Database issue):D432-8
pubmed: 25392411
Nucleic Acids Res. 2019 May 21;47(9):e50
pubmed: 30854567
Exp Mol Med. 2017 Mar 24;49(3):e308
pubmed: 28336957
ACS Chem Biol. 2018 Jun 15;13(6):1686-1694
pubmed: 29792670
J Biol Chem. 2015 Sep 4;290(36):21773-86
pubmed: 26088137
Front Immunol. 2018 Mar 01;9:329
pubmed: 29545792
PLoS One. 2019 Feb 11;14(2):e0212048
pubmed: 30742664
Neural Comput. 2000 Oct;12(10):2451-71
pubmed: 11032042
PLoS One. 2015 Jun 05;10(6):e0129125
pubmed: 26046845
Methods. 2013 Mar 15;60(1):99-110
pubmed: 23500657
Nat Biotechnol. 2010 Sep;28(9):965-9
pubmed: 20802495
Curr Opin Biotechnol. 2019 Dec;60:153-158
pubmed: 30849700
Front Immunol. 2018 Oct 16;9:2278
pubmed: 30386328
Cancer Discov. 2020 Aug 25;:
pubmed: 32847940
Bioinformatics. 2018 Dec 1;34(23):4138
pubmed: 29933431
MAbs. 2018 Apr;10(3):431-443
pubmed: 29376776
Nat Methods. 2018 Oct;15(10):816-822
pubmed: 30250057
Sci Rep. 2016 Aug 02;6:30312
pubmed: 27481573
Neural Comput. 1997 Nov 15;9(8):1735-80
pubmed: 9377276
J Mol Biol. 1991 Dec 5;222(3):581-97
pubmed: 1748994
Nat Med. 2015 Jan;21(1):86-91
pubmed: 25501908
Genomics. 2017 Oct;109(5-6):419-431
pubmed: 28669847
Bioinformatics. 2020 Apr 1;36(7):2126-2133
pubmed: 31778140
Front Immunol. 2015 Jan 12;5:673
pubmed: 25628622
MAbs. 2016 Oct;8(7):1177-1194
pubmed: 27416017
Methods. 2013 Mar 15;60(1):38-45
pubmed: 23276752
Comb Chem High Throughput Screen. 2003 Aug;6(5):421-32
pubmed: 12871049
Proc Natl Acad Sci U S A. 1996 Oct 1;93(20):10753-7
pubmed: 8855252