Novel multi-omics deconfounding variational autoencoders can obtain meaningful disease subtyping

Abstract Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Briefings in bioinformatics Jg. 25; H. 6; S. 512
Hauptverfasser: Li, Zuqi, Katz, Sonja, Saccenti, Edoardo, Fardo, David W, Claes, Peter, Martins dos Santos, Vitor A P, Van Steen, Kristel, Roshchupkin, Gennady V
Format: Journal Article
Sprache:Englisch
Veröffentlicht: England Oxford University Press 23.09.2024
Oxford Publishing Limited (England)
Schlagworte:
ISSN:1467-5463, 1477-4054, 1477-4054
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Abstract Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders—external factors unrelated to the condition, e.g. batch effect or age—on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include (i) removal of latent features correlated with confounders, (ii) a conditional VAE, (iii) adversarial training, and (iv) adding a regularization term to the loss function. Using real-life multi-omics data from The Cancer Genome Atlas, we simulated various confounding effects (linear, nonlinear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.
AbstractList Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders-external factors unrelated to the condition, e.g. batch effect or age-on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include (i) removal of latent features correlated with confounders, (ii) a conditional VAE, (iii) adversarial training, and (iv) adding a regularization term to the loss function. Using real-life multi-omics data from The Cancer Genome Atlas, we simulated various confounding effects (linear, nonlinear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.
Abstract Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders—external factors unrelated to the condition, e.g. batch effect or age—on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include (i) removal of latent features correlated with confounders, (ii) a conditional VAE, (iii) adversarial training, and (iv) adding a regularization term to the loss function. Using real-life multi-omics data from The Cancer Genome Atlas, we simulated various confounding effects (linear, nonlinear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.
Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders-external factors unrelated to the condition, e.g. batch effect or age-on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include (i) removal of latent features correlated with confounders, (ii) a conditional VAE, (iii) adversarial training, and (iv) adding a regularization term to the loss function. Using real-life multi-omics data from The Cancer Genome Atlas, we simulated various confounding effects (linear, nonlinear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders-external factors unrelated to the condition, e.g. batch effect or age-on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include (i) removal of latent features correlated with confounders, (ii) a conditional VAE, (iii) adversarial training, and (iv) adding a regularization term to the loss function. Using real-life multi-omics data from The Cancer Genome Atlas, we simulated various confounding effects (linear, nonlinear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.
Author Katz, Sonja
Van Steen, Kristel
Martins dos Santos, Vitor A P
Roshchupkin, Gennady V
Claes, Peter
Li, Zuqi
Saccenti, Edoardo
Fardo, David W
Author_xml – sequence: 1
  givenname: Zuqi
  surname: Li
  fullname: Li, Zuqi
– sequence: 2
  givenname: Sonja
  surname: Katz
  fullname: Katz, Sonja
– sequence: 3
  givenname: Edoardo
  surname: Saccenti
  fullname: Saccenti, Edoardo
– sequence: 4
  givenname: David W
  surname: Fardo
  fullname: Fardo, David W
– sequence: 5
  givenname: Peter
  surname: Claes
  fullname: Claes, Peter
– sequence: 6
  givenname: Vitor A P
  surname: Martins dos Santos
  fullname: Martins dos Santos, Vitor A P
– sequence: 7
  givenname: Kristel
  surname: Van Steen
  fullname: Van Steen, Kristel
– sequence: 8
  givenname: Gennady V
  surname: Roshchupkin
  fullname: Roshchupkin, Gennady V
  email: g.roshchupkin@erasmusmc.nl
BackLink https://www.ncbi.nlm.nih.gov/pubmed/39413796$$D View this record in MEDLINE/PubMed
BookMark eNp9kstrVTEQxoNU7ENX7uWAIIKcNpOc50qk-CiUutF1zGPONSUnuSYnF_rfm_uo2CJukmHy-75hMnNKjnzwSMhLoOdAR36hrLpQSmIL7Ak5gabv64a2zdE27vq6bTp-TE5TuqWU0X6AZ-SYjw3wfuxOyI-bsEFXzdkttg6z1akyqIOfQvbG-lW1kdHKxQYvXSXzEtDrYDCmSktfBbVI66sZpS_slF1lbEKZsEpZLXfrknxOnk7SJXxxuM_I908fv11-qa-_fr66_HBd62Zol3JSTTvTtt3IRwSKHBBY06PRrRwpG4yZuGGy6VsuDVA5KaBqADMYasau52fk_d53ndVcVOiXKJ1YRzvLeCeCtOLhi7c_xSpsBEAzcOBjceB7B2dxhSJEZcWG7ZS7OLuVkFooFIx1g-CMd7Ct-_ZQN4ZfGdMiZps0Oic9hpwEB-gLR3cFXj9Cb0OO5WO3FBsoQDt0hXr1dyN_OrifWQFgD-gYUoo4CW2X3YhKX9YJoGK7F6LshTjsRdG8e6S5t_03_WZPh7z-L_gbVF_JMA
CitedBy_id crossref_primary_10_3389_fmed_2025_1630788
crossref_primary_10_1016_j_imu_2025_101679
Cites_doi 10.1038/s41467-023-38125-0
10.3389/fgene.2021.772298
10.1002/cam4.1277
10.1093/bioinformatics/btaa976
10.1007/s10654-016-0155-5
10.1038/s41514-022-00085-y
10.1371/journal.pcbi.1009826
10.1158/2159-8290.CD-12-0095
10.1016/j.patcog.2018.12.015
10.1371/journal.pone.0210236
10.1109/BIBM47256.2019.8983228
10.1038/nmeth.4236
10.1136/bmj.k134
10.1038/ng.2764
10.1007/978-1-4939-9744-2_16
10.1016/j.cell.2018.03.022
10.1093/bioinformatics/btaa796
10.1007/978-3-030-87240-3_78
10.1038/s42256-022-00541-0
10.1016/j.tibtech.2017.02.012
10.1016/S0140-6736(16)30512-8
10.1038/s41598-020-65119-5
10.1093/gigascience/giac014
10.1093/ije/dyab274
10.1002/cncr.25936
10.1038/s41467-017-00289-x
10.3389/fgene.2019.01205
10.1093/nar/gkv1507
ContentType Journal Article
Copyright The Author(s) 2024. Published by Oxford University Press. 2024
The Author(s) 2024. Published by Oxford University Press.
Copyright_xml – notice: The Author(s) 2024. Published by Oxford University Press. 2024
– notice: The Author(s) 2024. Published by Oxford University Press.
DBID TOX
AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7QO
7SC
8FD
FR3
JQ2
K9.
L7M
L~C
L~D
P64
RC3
7X8
JLOSS
Q33
5PM
DOI 10.1093/bib/bbae512
DatabaseName Oxford Journals Open Access Collection
CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
Biotechnology Research Abstracts
Computer and Information Systems Abstracts
Technology Research Database
Engineering Research Database
ProQuest Computer Science Collection
ProQuest Health & Medical Complete (Alumni)
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
Biotechnology and BioEngineering Abstracts
Genetics Abstracts
MEDLINE - Academic
Université de Liège - Open Repository and Bibliography (ORBI) (Open Access titles only)
Université de Liège - Open Repository and Bibliography (ORBI)
PubMed Central (Full Participant titles)
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
Genetics Abstracts
Biotechnology Research Abstracts
Technology Research Database
Computer and Information Systems Abstracts – Academic
ProQuest Computer Science Collection
Computer and Information Systems Abstracts
ProQuest Health & Medical Complete (Alumni)
Engineering Research Database
Advanced Technologies Database with Aerospace
Biotechnology and BioEngineering Abstracts
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitleList MEDLINE
CrossRef

Genetics Abstracts
MEDLINE - Academic

Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: TOX
  name: Oxford Journals Open Access Collection
  url: https://academic.oup.com/journals/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Biology
EISSN 1477-4054
ExternalDocumentID PMC11483139
oai_orbi_ulg_ac_be_2268_323617
39413796
10_1093_bib_bbae512
10.1093/bib/bbae512
Genre Journal Article
GrantInformation_xml – fundername: Personalized Medicine in Infections: from Systems Biomedicine and Immunometabolism to Precision Diagnosis and Stratification Permitting Individualized Therapies
  grantid: 456008002
– fundername: Marie Sklodowska-Curie
  grantid: 860895
– fundername: The Netherlands Organisation for Health Research and Development
– fundername: NIA NIH HHS
  grantid: P30 AG072946
– fundername: PerMed Joint Transnational call JTC 2018
– fundername: NIA NIH HHS
  grantid: RF1 AG082339
– fundername: ;
– fundername: ;
  grantid: 456008002
– fundername: ;
  grantid: 860895
GroupedDBID ---
-E4
.2P
.I3
0R~
1TH
23N
2WC
36B
4.4
48X
53G
5GY
5VS
6J9
70D
8VB
AAGQS
AAHBH
AAIJN
AAIMJ
AAJKP
AAJQQ
AAMDB
AAMVS
AAOGV
AAPQZ
AAPXW
AARHZ
AAUQX
AAVAP
AAVLN
ABDBF
ABEJV
ABEUO
ABGNP
ABIXL
ABNKS
ABPQP
ABPTD
ABQLI
ABQTQ
ABWST
ABXVV
ABXZS
ABZBJ
ACGFO
ACGFS
ACGOD
ACIWK
ACPRK
ACUFI
ACUHS
ACUXJ
ACYTK
ADBBV
ADEYI
ADFTL
ADGKP
ADGZP
ADHKW
ADHZD
ADOCK
ADPDF
ADQBN
ADRDM
ADRTK
ADVEK
ADYVW
ADZTZ
ADZXQ
AECKG
AEGPL
AEGXH
AEJOX
AEKKA
AEKSI
AELWJ
AEMDU
AEMOZ
AENEX
AENZO
AEPUE
AETBJ
AEWNT
AFFZL
AFGWE
AFIYH
AFOFC
AFRAH
AGINJ
AGKEF
AGQXC
AGSYK
AHMBA
AHQJS
AHXPO
AIAGR
AIJHB
AJEEA
AJEUX
AKHUL
AKVCP
AKWXX
ALMA_UNASSIGNED_HOLDINGS
ALTZX
ALUQC
ALXQX
AMNDL
ANAKG
APIBT
APWMN
ARIXL
AXUDD
AYOIW
AZVOD
BAWUL
BAYMD
BEYMZ
BHONS
BQDIO
BQUQU
BSWAC
BTQHN
C1A
C45
CAG
CDBKE
COF
CS3
CZ4
DAKXR
DIK
DILTD
DU5
D~K
E3Z
EAD
EAP
EAS
EBA
EBC
EBD
EBR
EBS
EBU
EE~
EJD
EMB
EMK
EMOBN
EST
ESX
F5P
F9B
FHSFR
FLIZI
FLUFQ
FOEOM
FQBLK
GAUVT
GJXCC
GROUPED_DOAJ
GX1
H13
H5~
HAR
HW0
HZ~
IOX
J21
JXSIZ
K1G
KBUDW
KOP
KSI
KSN
M-Z
M49
MK~
ML0
N9A
NGC
NLBLG
NMDNZ
NOMLY
NU-
O0~
O9-
OAWHX
ODMLO
OJQWA
OK1
OVD
OVEED
P2P
PAFKI
PEELM
PQQKQ
Q1.
Q5Y
QWB
RD5
RPM
RUSNO
RW1
RXO
SV3
TEORI
TH9
TJP
TLC
TOX
TR2
TUS
W8F
WOQ
X7H
YAYTL
YKOAZ
YXANX
ZKX
ZL0
~91
77I
AAYXX
AHGBF
CITATION
ROX
CGR
CUY
CVF
ECM
EIF
NPM
7QO
7SC
8FD
FR3
JQ2
K9.
L7M
L~C
L~D
P64
RC3
7X8
JLOSS
Q33
5PM
ID FETCH-LOGICAL-c485t-c40c06d556939e10e31e1247edc5a9028ddf3d2a4753ad10afb10b81d8d0d9673
IEDL.DBID TOX
ISICitedReferencesCount 2
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001332424100005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1467-5463
1477-4054
IngestDate Tue Nov 04 02:04:55 EST 2025
Sat Nov 29 01:28:32 EST 2025
Sat Sep 27 20:21:48 EDT 2025
Mon Oct 06 16:56:50 EDT 2025
Wed Jul 23 01:46:56 EDT 2025
Sat Nov 29 04:20:24 EST 2025
Tue Nov 18 22:22:52 EST 2025
Wed Apr 02 07:03:59 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 6
Keywords fairness
multi-omics
deep learning
clustering
autoencoder
confounders
Language English
License This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
https://creativecommons.org/licenses/by/4.0
The Author(s) 2024. Published by Oxford University Press.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c485t-c40c06d556939e10e31e1247edc5a9028ddf3d2a4753ad10afb10b81d8d0d9673
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
scopus-id:2-s2.0-85206648620
Zuqi Li and Sonja Katz contributed equally.
ORCID 0000-0001-9868-5033
OpenAccessLink https://dx.doi.org/10.1093/bib/bbae512
PMID 39413796
PQID 3128011586
PQPubID 26846
ParticipantIDs pubmedcentral_primary_oai_pubmedcentral_nih_gov_11483139
liege_orbi_v2_oai_orbi_ulg_ac_be_2268_323617
proquest_miscellaneous_3117617039
proquest_journals_3128011586
pubmed_primary_39413796
crossref_citationtrail_10_1093_bib_bbae512
crossref_primary_10_1093_bib_bbae512
oup_primary_10_1093_bib_bbae512
PublicationCentury 2000
PublicationDate 2024-Sep-23
PublicationDateYYYYMMDD 2024-09-23
PublicationDate_xml – month: 09
  year: 2024
  text: 2024-Sep-23
  day: 23
PublicationDecade 2020
PublicationPlace England
PublicationPlace_xml – name: England
– name: Oxford
PublicationTitle Briefings in bioinformatics
PublicationTitleAlternate Brief Bioinform
PublicationYear 2024
Publisher Oxford University Press
Oxford Publishing Limited (England)
Publisher_xml – name: Oxford University Press
– name: Oxford Publishing Limited (England)
References Čuklina (2024101622180509600_ref3) 2020
Zhang (2024101622180509600_ref31) 2019
Simidjievski (2024101622180509600_ref1); 10
Cerami (2024101622180509600_ref19) 2012; 2
Chen (2024101622180509600_ref17) 2017; 8
Rodriguez (2024101622180509600_ref37) 2019; 14
Hoadley (2024101622180509600_ref18) 2018; 173
Weinstein (2024101622180509600_ref15) 2013; 45
González-Reymúndez (2024101622180509600_ref30) 2020; 10
Yu (2024101622180509600_ref33) 2022; 18
de Lima Camillo (2024101622180509600_ref10) 2022; 8
Pourhoseingholi (2024101622180509600_ref4) 2012; 5
Kartsonaki (2024101622180509600_ref13) 2022; 51
Liu (2024101622180509600_ref9)
Lawry Aguila (2024101622180509600_ref6) 2022
Kiselev (2024101622180509600_ref25) 2017; 14
He (2024101622180509600_ref36) 2022; 4
Kuzub (2024101622180509600_ref11) 2022; 12
Wang (2024101622180509600_ref32) 2023
Radhakrishnan (2024101622180509600_ref5) 2023; 14
Horne (2024101622180509600_ref28) 2017; 37
Falcon (2024101622180509600_ref26) 2019
de Lima Camillo (2024101622180509600_ref20) 2022; 8
Goh (2024101622180509600_ref2) 2017; 35
Bahrami (2024101622180509600_ref8) 2021; 37
Dincer (2024101622180509600_ref7) 2020; 36
Odegaard (2024101622180509600_ref14) 2011; 117
Kamat (2024101622180509600_ref27) 2016; 388
Adeli (2024101622180509600_ref34) 2021
Fan (2024101622180509600_ref24) 2019; 88
Uyar (2024101622180509600_ref29) 2021
Sohn (2024101622180509600_ref23) 2015
Tu (2024101622180509600_ref12) 2018; 360
Li (2024101622180509600_ref21) 2016; 31
Colaprico (2024101622180509600_ref16) 2016; 44
Nipp (2024101622180509600_ref22) 2018; 7
Chyzhyk (2024101622180509600_ref35) 2022; 11
References_xml – volume: 14
  start-page: 2436
  year: 2023
  ident: 2024101622180509600_ref5
  article-title: Cross-modal autoencoder framework learns holistic representations of cardiovascular state
  publication-title: Nat Commun
  doi: 10.1038/s41467-023-38125-0
– volume: 12
  year: 2022
  ident: 2024101622180509600_ref11
  article-title: Evaluation of epigenetic age based on dna methylation analysis of several cpg sites in ukrainian population
  publication-title: Front Genet
  doi: 10.3389/fgene.2021.772298
– volume: 7
  start-page: 525
  year: 2018
  ident: 2024101622180509600_ref22
  article-title: Disparities in cancer outcomes across age, sex, and race/ethnicity among patients with pancreatic cancer
  publication-title: Cancer Med
  doi: 10.1002/cam4.1277
– volume: 37
  start-page: 1345
  year: 2021
  ident: 2024101622180509600_ref8
  article-title: Deep feature extraction of single-cell transcriptomes by generative adversarial network
  publication-title: Bioinformatics
  doi: 10.1093/bioinformatics/btaa976
– volume: 31
  start-page: 603
  year: 2016
  ident: 2024101622180509600_ref21
  article-title: How much do tumor stage and treatment explain socioeconomic inequalities in breast cancer survival? Applying causal mediation analysis to population-based data
  publication-title: Eur J Epidemiol
  doi: 10.1007/s10654-016-0155-5
– volume: 8
  start-page: 1
  year: 2022
  ident: 2024101622180509600_ref20
  article-title: A pan-tissue DNA-methylation epigenetic clock based on deep learning
  publication-title: npj Aging
  doi: 10.1038/s41514-022-00085-y
– volume: 18
  start-page: e1009826
  year: 2022
  ident: 2024101622180509600_ref33
  article-title: AIME: Autoencoder-based integrative multi-omics data embedding that allows for confounder adjustments
  publication-title: PLoS Comput Biol
  doi: 10.1371/journal.pcbi.1009826
– volume: 2
  start-page: 401
  year: 2012
  ident: 2024101622180509600_ref19
  article-title: The cbio cancer genomics portal: An open platform for exploring multidimensional cancer genomics data
  publication-title: Cancer Discov
  doi: 10.1158/2159-8290.CD-12-0095
– volume: 5
  start-page: 79
  year: 2012
  ident: 2024101622180509600_ref4
  article-title: How to control confounding effects by statistical analysis
  publication-title: Gastroenterol Hepatol Bed Bench
– volume: 88
  start-page: 643
  year: 2019
  ident: 2024101622180509600_ref24
  article-title: Autoencoder node saliency: Selecting relevant latent representations
  publication-title: Pattern Recognit
  doi: 10.1016/j.patcog.2018.12.015
– start-page: 2513
  volume-title: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision
  year: 2021
  ident: 2024101622180509600_ref34
  article-title: Representation learning with statistical independence to mitigate bias
– volume: 14
  start-page: e0210236
  year: 2019
  ident: 2024101622180509600_ref37
  article-title: Clustering algorithms: A comparative approach
  publication-title: PloS One
  doi: 10.1371/journal.pone.0210236
– start-page: 430
  volume-title: Medical Image Computing and Computer Assisted Intervention—MICCAI 2022, Lecture Notes in Computer Science
  year: 2022
  ident: 2024101622180509600_ref6
  article-title: Conditional VAEs for confound removal and normative modelling of neurodegenerative diseases
– start-page: 765
  volume-title: 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
  year: 2019
  ident: 2024101622180509600_ref31
  article-title: Integrated multi-omics analysis using variational autoencoders: Application to pan-cancer classification
  doi: 10.1109/BIBM47256.2019.8983228
– volume: 8
  start-page: 4
  year: 2022
  ident: 2024101622180509600_ref10
  article-title: A pan-tissue DNA-methylation epigenetic clock based on deep learning
  publication-title: npj Aging
  doi: 10.1038/s41514-022-00085-y
– volume: 14
  year: 2017
  ident: 2024101622180509600_ref25
  article-title: Sc3: Consensus clustering of single-cell rna-seq data
  publication-title: Nat Methods
  doi: 10.1038/nmeth.4236
– year: 2019
  ident: 2024101622180509600_ref26
  article-title: PyTorch lightning
– volume: 37
  start-page: 2785
  year: 2017
  ident: 2024101622180509600_ref28
  article-title: Cancer tissue classification, associated therapeutic implications and PDT as an alternative
  publication-title: Anticancer Res
– volume: 360
  start-page: k134
  year: 2018
  ident: 2024101622180509600_ref12
  article-title: Cancer risk associated with chronic diseases and disease markers: Prospective cohort study
  publication-title: BMJ
  doi: 10.1136/bmj.k134
– volume: 45
  start-page: 1113
  year: 2013
  ident: 2024101622180509600_ref15
  article-title: The cancer genome atlas pan-cancer analysis project
  publication-title: Nat Genet
  doi: 10.1038/ng.2764
– start-page: 373
  volume-title: Mass Spectrometry Data Analysis in Proteomics, Methods in Molecular Biology
  year: 2020
  ident: 2024101622180509600_ref3
  article-title: Review of batch effects prevention, diagnostics, and correction approaches
  doi: 10.1007/978-1-4939-9744-2_16
– volume: 173
  start-page: 291
  year: 2018
  ident: 2024101622180509600_ref18
  article-title: Cell-of-origin patterns dominate the molecular classification of 10,000 tumors from 33 types of cancer
  publication-title: Cell
  doi: 10.1016/j.cell.2018.03.022
– volume: 36
  start-page: i573
  year: 2020
  ident: 2024101622180509600_ref7
  article-title: Adversarial deconfounding autoencoder for learning robust gene expression embeddings
  publication-title: Bioinformatics
  doi: 10.1093/bioinformatics/btaa796
– volume-title: Medical Image Computing and Computer Assisted Intervention – MICCAI 2021, Strasbourg, France, 2021. Proceedings, Part V 24. Springer International Publishing, New York City, United States.
  ident: 2024101622180509600_ref9
  article-title: Projection-wise disentangling for fair and interpretable representation learning: Application to 3d facial shape analysis
  doi: 10.1007/978-3-030-87240-3_78
– volume: 4
  start-page: 879
  year: 2022
  ident: 2024101622180509600_ref36
  article-title: A context-aware deconfounding autoencoder for robust prediction of personalized clinical drug response from cell-line compound screening
  publication-title: Nat Mach Intell
  doi: 10.1038/s42256-022-00541-0
– volume: 35
  start-page: 498
  year: 2017
  ident: 2024101622180509600_ref2
  article-title: Why batch effects matter in omics data, and how to avoid them
  publication-title: Trends Biotechnol
  doi: 10.1016/j.tibtech.2017.02.012
– volume: 388
  start-page: 2796
  year: 2016
  ident: 2024101622180509600_ref27
  article-title: Bladder cancer
  publication-title: Lancet
  doi: 10.1016/S0140-6736(16)30512-8
– volume: 10
  start-page: 8341
  year: 2020
  ident: 2024101622180509600_ref30
  article-title: Multi-omic signatures identify pan-cancer classes of tumors beyond tissue of origin
  publication-title: Sci Rep
  doi: 10.1038/s41598-020-65119-5
– volume: 11
  start-page: giac014
  year: 2022
  ident: 2024101622180509600_ref35
  article-title: How to remove or control confounds in predictive models, with applications to brain biomarkers
  publication-title: GigaScience
  doi: 10.1093/gigascience/giac014
– start-page: 1
  volume-title: 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI)
  year: 2023
  ident: 2024101622180509600_ref32
  article-title: Normative Modeling via conditional Variational autoencoder and adversarial learning to identify brain dysfunction in Alzheimer’s disease
– volume: 51
  start-page: 817
  year: 2022
  ident: 2024101622180509600_ref13
  article-title: Circulating proteins and risk of pancreatic cancer: A case-subcohort study among Chinese adults
  publication-title: Int J Epidemiol
  doi: 10.1093/ije/dyab274
– volume: 117
  start-page: 3841
  year: 2011
  ident: 2024101622180509600_ref14
  article-title: Body mass index and risk of colorectal cancer in chinese singaporeans: The Singapore chinese health study
  publication-title: Cancer
  doi: 10.1002/cncr.25936
– volume: 8
  start-page: 199
  year: 2017
  ident: 2024101622180509600_ref17
  article-title: Pan-urologic cancer genomic subtypes that transcend tissue of origin
  publication-title: Nat Commun
  doi: 10.1038/s41467-017-00289-x
– start-page: 2021
  year: 2021
  ident: 2024101622180509600_ref29
  article-title: Multi-omics and deep learning provide a multifaceted view of cancer
  publication-title: bioRxiv
– volume: 10
  ident: 2024101622180509600_ref1
  article-title: Variational autoencoders for cancer data integration: Design principles and computational practice
  publication-title: Front Genet
  doi: 10.3389/fgene.2019.01205
– volume-title: Advances in Neural Information Processing Systems
  year: 2015
  ident: 2024101622180509600_ref23
  article-title: Learning structured output representation using deep conditional generative models
– volume: 44
  start-page: e71
  year: 2016
  ident: 2024101622180509600_ref16
  article-title: TCGAbiolinks: An R/Bioconductor package for integrative analysis of TCGA data
  publication-title: Nucleic Acids Res
  doi: 10.1093/nar/gkv1507
SSID ssj0020781
Score 2.4273796
Snippet Abstract Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of...
Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of...
SourceID pubmedcentral
liege
proquest
pubmed
crossref
oup
SourceType Open Access Repository
Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 512
SubjectTerms Algorithms
Biological analysis
Biological effects
Cancer
Cluster Analysis
Clustering
Computational Biology - methods
Data integration
Deep Learning
Genomics - methods
Heterogeneity
Humans
Life sciences
Machine learning
Multiomics
Neoplasms - classification
Neoplasms - genetics
Precision medicine
Problem Solving Protocol
Regularization
Sciences du vivant
Training
Unsupervised learning
Unsupervised Machine Learning
Title Novel multi-omics deconfounding variational autoencoders can obtain meaningful disease subtyping
URI https://www.ncbi.nlm.nih.gov/pubmed/39413796
https://www.proquest.com/docview/3128011586
https://www.proquest.com/docview/3117617039
https://orbi.uliege.be/handle/2268/323617
https://pubmed.ncbi.nlm.nih.gov/PMC11483139
Volume 25
WOSCitedRecordID wos001332424100005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 1477-4054
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0020781
  issn: 1477-4054
  databaseCode: DOA
  dateStart: 20240101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVASL
  databaseName: Oxford Journals Open Access Collection
  customDbUrl:
  eissn: 1477-4054
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0020781
  issn: 1477-4054
  databaseCode: TOX
  dateStart: 20000101
  isFulltext: true
  titleUrlDefault: https://academic.oup.com/journals/
  providerName: Oxford University Press
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1Lb9QwELaggMSFdyFQFiP1hIgax0lsHxFqxWnhUKS9Gb8CK6VJtUlW6r9nJkmj7qqCXqJEthPHD803nplvCDn2ApQCnhdxKIyJYSey2MDsxkGaImNeWXgakk2I5VKuVurH5CDb3mLCV_zEru2JtSbkQzJhlktMVHD-fTXrVchXMwYRiRjZ3acwvL22O4LnQYUm6b2gthvYct9F8obMOXt6194-I08mVEm_jMvgObkX6hfk0Zhn8uol-bVstqGig_NgjGHILfWoCJeYUwlkF92CxjydClLTdw2yW6KHM4WBp43F4wN6EQyeoZR9RSerDm17211hwNUr8vPs9Pzrt3hKrRC7TOYdXBOXFD7PC8VVYEngLICkF_DDuUFCF-9L7lOTgTZjPEtMaVliAdtKn3hVCH5IDuqmDm8I9alznJc2y4TNPOBNqwygMBkcKxIvXUQ-XY-7dhPvOKa_qPRo_-YahkxPQxaR47ny5Ui3cXu1z8ME6mZj13qbaiTJHu776rc2TtugAVdKzZFbRkTkA8zzv194dL0G9LSJW81BdiNilkVEPs7FsP3QpmLq0PRYhwnktOcqIq_HJTN_hytACEJBa7mzmOYK2Ovdknr9Z6D4Ri2VAzh_-9-evyOPU8BZ6MKS8iNy0G368J48dNtu3W4W5L5YycVw0LAYNs1f8fAUmA
linkProvider Oxford University Press
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Novel+multi-omics+deconfounding+variational+autoencoders+can+obtain+meaningful+disease+subtyping&rft.jtitle=Briefings+in+bioinformatics&rft.au=Li%2C+Zuqi&rft.au=Katz%2C+Sonja&rft.au=Saccenti%2C+Edoardo&rft.au=Fardo%2C+David+W&rft.date=2024-09-23&rft.issn=1477-4054&rft.volume=25&rft.issue=6&rft.spage=512&rft_id=info:doi/10.1093%2Fbib%2Fbbae512&rft_id=info%3Apmid%2F39413796&rft.externalDBID=n%2Fa&rft.externalDocID=oai_orbi_ulg_ac_be_2268_323617
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1467-5463&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1467-5463&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1467-5463&client=summon