DISCERN: deep single-cell expression reconstruction for improved cell clustering and cell subtype and state detection.
Auto encoder
Batch effect correction
COVID-19
Cell clustering
Cell type identification
Deep Learning
Expression reconstruction
Imputation
Machine Learning
Probabilistic modeling
RNA sequencing
Reference atlas mapping
Single-cell RNA-seq
T helper cell
Transcription factor analysis
Transfer learning
Journal
Genome biology
ISSN: 1474-760X
Titre abrégé: Genome Biol
Pays: England
ID NLM: 100960660
Informations de publication
Date de publication:
20 09 2023
20 09 2023
Historique:
received:
15
11
2022
accepted:
23
08
2023
medline:
22
9
2023
pubmed:
21
9
2023
entrez:
20
9
2023
Statut:
epublish
Résumé
Single-cell sequencing provides detailed insights into biological processes including cell differentiation and identity. While providing deep cell-specific information, the method suffers from technical constraints, most notably a limited number of expressed genes per cell, which leads to suboptimal clustering and cell type identification. Here, we present DISCERN, a novel deep generative network that precisely reconstructs missing single-cell gene expression using a reference dataset. DISCERN outperforms competing algorithms in expression inference resulting in greatly improved cell clustering, cell type and activity detection, and insights into the cellular regulation of disease. We show that DISCERN is robust against differences between batches and is able to keep biological differences between batches, which is a common problem for imputation and batch correction algorithms. We use DISCERN to detect two unseen COVID-19-associated T cell types, cytotoxic CD4 Thus, DISCERN is a flexible tool for reconstructing missing single-cell gene expression using a reference dataset and can easily be applied to a variety of data sets yielding novel insights, e.g., into disease mechanisms.
Sections du résumé
BACKGROUND
Single-cell sequencing provides detailed insights into biological processes including cell differentiation and identity. While providing deep cell-specific information, the method suffers from technical constraints, most notably a limited number of expressed genes per cell, which leads to suboptimal clustering and cell type identification.
RESULTS
Here, we present DISCERN, a novel deep generative network that precisely reconstructs missing single-cell gene expression using a reference dataset. DISCERN outperforms competing algorithms in expression inference resulting in greatly improved cell clustering, cell type and activity detection, and insights into the cellular regulation of disease. We show that DISCERN is robust against differences between batches and is able to keep biological differences between batches, which is a common problem for imputation and batch correction algorithms. We use DISCERN to detect two unseen COVID-19-associated T cell types, cytotoxic CD4
CONCLUSIONS
Thus, DISCERN is a flexible tool for reconstructing missing single-cell gene expression using a reference dataset and can easily be applied to a variety of data sets yielding novel insights, e.g., into disease mechanisms.
Identifiants
pubmed: 37730638
doi: 10.1186/s13059-023-03049-x
pii: 10.1186/s13059-023-03049-x
pmc: PMC10510283
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
212Informations de copyright
© 2023. BioMed Central Ltd., part of Springer Nature.
Références
BMC Bioinformatics. 2013 Apr 15;14:128
pubmed: 23586463
Nat Immunol. 2021 Jun;22(6):735-745
pubmed: 34017124
Genome Biol. 2019 May 6;20(1):88
pubmed: 31060596
Nat Biotechnol. 2020 Jun;38(6):737-746
pubmed: 32341560
Nat Methods. 2022 Jan;19(1):41-50
pubmed: 34949812
Front Immunol. 2018 Nov 13;9:2605
pubmed: 30555458
BMC Genomics. 2018 Jun 19;19(1):477
pubmed: 29914354
Bioinformatics. 2020 Dec 30;36(Suppl_2):i610-i617
pubmed: 33381839
Sci Immunol. 2019 Jul 5;4(37):
pubmed: 31278120
Adipocyte. 2014 Jan 1;3(1):58-62
pubmed: 24575371
Nat Biotechnol. 2022 Feb;40(2):163-166
pubmed: 35132262
Respir Med. 2011 Nov;105(11):1648-54
pubmed: 21763119
Immunology. 2021 May;163(1):3-18
pubmed: 33064842
Front Immunol. 2020 Mar 12;11:348
pubmed: 32226427
Cell Rep Methods. 2021 Dec 20;2(1):100133
pubmed: 35474868
Nat Biotechnol. 2022 Jan;40(1):121-130
pubmed: 34462589
Genomics Proteomics Bioinformatics. 2021 Apr;19(2):253-266
pubmed: 33662621
Nature. 2014 Jun 19;510(7505):363-9
pubmed: 24919153
Nat Commun. 2019 Jan 23;10(1):390
pubmed: 30674886
Int Immunopharmacol. 2020 Dec;89(Pt A):107071
pubmed: 33221703
Genome Res. 2017 Feb;27(2):208-222
pubmed: 27864352
Nat Med. 2021 May;27(5):904-916
pubmed: 33879890
Immunity. 2011 Apr 22;34(4):566-78
pubmed: 21511185
Nat Rev Immunol. 2019 Aug;19(8):476
pubmed: 31243349
Annu Rev Immunol. 2010;28:57-78
pubmed: 20307208
Hum Pathol. 2020 Sep;103:120-126
pubmed: 32702400
Database (Oxford). 2019 Jan 1;2019:
pubmed: 30951143
J Immunol. 2012 Jun 1;188(11):5438-47
pubmed: 22539793
Cell Syst. 2016 Oct 26;3(4):346-360.e4
pubmed: 27667365
J Allergy Clin Immunol. 2010 Aug;126(2):280-9, 289.e1-7
pubmed: 20624644
Nat Methods. 2017 Sep;14(9):865-868
pubmed: 28759029
NAR Genom Bioinform. 2020 Dec;2(4):lqaa077
pubmed: 33029585
Nat Commun. 2020 Jul 6;11(1):3434
pubmed: 32632085
PLoS One. 2011;6(11):e28011
pubmed: 22132193
Nat Methods. 2021 Aug;18(8):881-892
pubmed: 34282327
Genome Biol. 2020 Aug 6;21(1):196
pubmed: 32762710
Microbiol Immunol. 2018 May;62(5):348-356
pubmed: 29577371
Genome Biol. 2018 Feb 6;19(1):15
pubmed: 29409532
Nat Commun. 2018 Mar 8;9(1):997
pubmed: 29520097
Int Immunol. 2011 Jul;23(7):415-20
pubmed: 21632975
Cell. 2020 Nov 25;183(5):1340-1353.e16
pubmed: 33096020
Nat Methods. 2014 Jan;11(1):1
pubmed: 24524124
Mol Immunol. 2006 Mar;43(9):1497-507
pubmed: 16143398
J Immunol. 2017 Jun 15;198(12):4659-4671
pubmed: 28507030
Cell. 2021 May 27;184(11):3006-3021.e17
pubmed: 33930287
Clin Exp Pharmacol Physiol. 2016 Apr;43(4):410-6
pubmed: 26845249
Cell. 2018 Jul 26;174(3):716-729.e27
pubmed: 29961576
Nat Methods. 2019 Mar;16(3):243-245
pubmed: 30742040
Sci Immunol. 2021 Feb 23;6(56):
pubmed: 33622974
Nat Methods. 2017 Nov;14(11):1083-1086
pubmed: 28991892
Int Immunol. 2004 Aug;16(8):1109-24
pubmed: 15210650
Eur J Immunol. 2015 Feb;45(2):474-9
pubmed: 25446972
PLoS One. 2011;6(7):e22560
pubmed: 21829468
J Mol Cell Biol. 2021 Jul 6;13(3):197-209
pubmed: 33751111
BMC Cancer. 2015 Oct 16;15:717
pubmed: 26474968
Immunity. 2008 Jan;28(1):29-39
pubmed: 18164222
Int J Womens Health. 2021 Oct 22;13:991-1004
pubmed: 34712062
Nat Commun. 2020 Jan 9;11(1):166
pubmed: 31919373
Genome Res. 2021 Oct;31(10):1753-1766
pubmed: 34035047
Cell Tissue Res. 2021 Aug;385(2):435-443
pubmed: 34125286
J Immunol. 2021 Jun 1;206(11):2714-2724
pubmed: 34011519
Front Immunol. 2018 Nov 29;9:2788
pubmed: 30555473
Nat Commun. 2020 Dec 11;11(1):6357
pubmed: 33311473
Proc Natl Acad Sci U S A. 2005 Oct 25;102(43):15545-50
pubmed: 16199517
Nat Commun. 2018 Jan 18;9(1):284
pubmed: 29348443
Cochrane Database Syst Rev. 2021 Aug 16;8:CD014963
pubmed: 34396514
J Cell Physiol. 2021 Apr;236(4):2829-2839
pubmed: 32926425
Vitam Horm. 2014;95:165-93
pubmed: 24559918
Nat Methods. 2019 Aug;16(8):715-721
pubmed: 31363220
F1000Res. 2018 Nov 2;7:1740
pubmed: 30906525
Eur J Immunol. 2013 Nov;43(11):2797-809
pubmed: 24258910
Curr Diabetes Rev. 2013 Jan 1;9(1):25-53
pubmed: 22974359
Cell Rep. 2016 Jul 12;16(2):392-404
pubmed: 27346359
Biostatistics. 2022 Dec 13;:
pubmed: 36511385
Oncoimmunology. 2015 Apr 14;4(8):e1026534
pubmed: 26405587
Scand J Immunol. 2003 Feb;57(2):192-8
pubmed: 12588667
Cell Syst. 2016 Oct 26;3(4):385-394.e3
pubmed: 27693023
Immunity. 2008 Jan;28(1):64-74
pubmed: 18191595
Nat Biotechnol. 2020 Jun;38(6):708-714
pubmed: 32518404
Sci Rep. 2019 Mar 26;9(1):5233
pubmed: 30914743
Cell. 2019 Jun 13;177(7):1888-1902.e21
pubmed: 31178118
Genome Biol. 2020 Aug 27;21(1):218
pubmed: 32854757
Cell Rep. 2019 Feb 5;26(6):1627-1640.e7
pubmed: 30726743
Bioinformatics. 2020 May 1;36(9):2778-2786
pubmed: 31971583
Eur J Immunol. 2019 Mar;49(3):398-412
pubmed: 30620397
Genome Biol. 2020 Feb 7;21(1):31
pubmed: 32033589
Nat Commun. 2019 Nov 28;10(1):5416
pubmed: 31780648
Genome Biol. 2022 Jan 21;23(1):31
pubmed: 35063006
Nat Immunol. 2019 Dec;20(12):1692-1699
pubmed: 31745340
Cell Metab. 2016 Oct 11;24(4):593-607
pubmed: 27667667
Nat Commun. 2017 Jan 16;8:14049
pubmed: 28091601
Mol Syst Biol. 2021 Jan;17(1):e9620
pubmed: 33491336
Int J Cancer. 2016 Sep 1;139(5):976-85
pubmed: 27012367
Cell Stem Cell. 2016 Aug 4;19(2):266-277
pubmed: 27345837
Mol Syst Biol. 2019 Jun 19;15(6):e8746
pubmed: 31217225
Nat Commun. 2022 Jan 11;13(1):192
pubmed: 35017482
Genome Biol. 2019 Oct 18;20(1):211
pubmed: 31627739
Eur J Immunol. 2019 Jan;49(1):38-41
pubmed: 30536524
NAR Genom Bioinform. 2021 Dec 22;3(4):lqab118
pubmed: 34988441
World J Surg Oncol. 2022 Jan 4;20(1):2
pubmed: 34980144
J Immunol. 2009 May 1;182(9):5702-11
pubmed: 19380817
Braz J Infect Dis. 2017 Mar - Apr;21(2):155-161
pubmed: 27932286
Genome Biol. 2023 Sep 20;24(1):212
pubmed: 37730638
Mol Cell Proteomics. 2016 Mar;15(3):1007-16
pubmed: 26637539
Front Immunol. 2020 Feb 14;11:206
pubmed: 32117317
JCI Insight. 2017 May 18;2(10):
pubmed: 28515365