Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation

The Cancer Genome Atlas Research Network

Research output: Contribution to journalArticle

78 Citations (Scopus)

Abstract

Cancer progression involves the gradual loss of a differentiated phenotype and acquisition of progenitor and stem-cell-like features. Here, we provide novel stemness indices for assessing the degree of oncogenic dedifferentiation. We used an innovative one-class logistic regression (OCLR) machine-learning algorithm to extract transcriptomic and epigenetic feature sets derived from non-transformed pluripotent stem cells and their differentiated progeny. Using OCLR, we were able to identify previously undiscovered biological mechanisms associated with the dedifferentiated oncogenic state. Analyses of the tumor microenvironment revealed unanticipated correlation of cancer stemness with immune checkpoint expression and infiltrating immune cells. We found that the dedifferentiated oncogenic phenotype was generally most prominent in metastatic tumors. Application of our stemness indices to single-cell data revealed patterns of intra-tumor molecular heterogeneity. Finally, the indices allowed for the identification of novel targets and possible targeted therapies aimed at tumor differentiation. Stemness features extracted from transcriptomic and epigenetic data from TCGA tumors reveal novel biological and clinical insight, as well as potential drug targets for anti-cancer therapies.

Original languageEnglish (US)
Pages (from-to)338-354.e15
JournalCell
Volume173
Issue number2
DOIs
StatePublished - Apr 5 2018

Fingerprint

Learning systems
Tumors
Neoplasms
Stem cells
Logistics
Epigenomics
Stem Cells
Logistic Models
Phenotype
Learning algorithms
Pluripotent Stem Cells
Tumor Microenvironment
Machine Learning
Pharmaceutical Preparations
Therapeutics

Keywords

  • cancer stem cells
  • dedifferentiation
  • epigenomic
  • genomic
  • machine learning
  • pan-cancer
  • stemness
  • The Cancer Genome Atlas

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation. / The Cancer Genome Atlas Research Network.

In: Cell, Vol. 173, No. 2, 05.04.2018, p. 338-354.e15.

Research output: Contribution to journalArticle

The Cancer Genome Atlas Research Network. / Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation. In: Cell. 2018 ; Vol. 173, No. 2. pp. 338-354.e15.
@article{256f11334428420e923aa612b9828a40,
title = "Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation",
abstract = "Cancer progression involves the gradual loss of a differentiated phenotype and acquisition of progenitor and stem-cell-like features. Here, we provide novel stemness indices for assessing the degree of oncogenic dedifferentiation. We used an innovative one-class logistic regression (OCLR) machine-learning algorithm to extract transcriptomic and epigenetic feature sets derived from non-transformed pluripotent stem cells and their differentiated progeny. Using OCLR, we were able to identify previously undiscovered biological mechanisms associated with the dedifferentiated oncogenic state. Analyses of the tumor microenvironment revealed unanticipated correlation of cancer stemness with immune checkpoint expression and infiltrating immune cells. We found that the dedifferentiated oncogenic phenotype was generally most prominent in metastatic tumors. Application of our stemness indices to single-cell data revealed patterns of intra-tumor molecular heterogeneity. Finally, the indices allowed for the identification of novel targets and possible targeted therapies aimed at tumor differentiation. Stemness features extracted from transcriptomic and epigenetic data from TCGA tumors reveal novel biological and clinical insight, as well as potential drug targets for anti-cancer therapies.",
keywords = "cancer stem cells, dedifferentiation, epigenomic, genomic, machine learning, pan-cancer, stemness, The Cancer Genome Atlas",
author = "{The Cancer Genome Atlas Research Network} and Malta, {Tathiane M.} and Artem Sokolov and Gentles, {Andrew J.} and Tomasz Burzykowski and Laila Poisson and Weinstein, {John N.} and Bożena Kamińska and Joerg Huelsken and Larsson Omberg and Olivier Gevaert and Antonio Colaprico and Patrycja Czerwińska and Sylwia Mazurek and Lopa Mishra and Holger Heyn and Alex Krasnitz and Godwin, {Andrew K.} and Lazar, {Alexander J.} and Caesar-Johnson, {Samantha J.} and Demchok, {John A.} and Ina Felau and Melpomeni Kasapi and Ferguson, {Martin L.} and Hutter, {Carolyn M.} and Sofia, {Heidi J.} and Roy Tarnuzzer and Zhining Wang and Liming Yang and Zenklusen, {Jean C.} and Zhang, {Jiashan (Julia)} and Sudha Chudamani and Jia Liu and Laxmi Lolla and Rashi Naresh and Todd Pihl and Qiang Sun and Yunhu Wan and Ye Wu and Juok Cho and Timothy DeFreitas and Scott Frazer and Nils Gehlenborg and Gad Getz and Heiman, {David I.} and Jaegil Kim and Lawrence, {Michael S.} and Pei Lin and Gordon Mills and Paul Spellman and George Thomas",
year = "2018",
month = "4",
day = "5",
doi = "10.1016/j.cell.2018.03.034",
language = "English (US)",
volume = "173",
pages = "338--354.e15",
journal = "Cell",
issn = "0092-8674",
publisher = "Cell Press",
number = "2",

}

TY - JOUR

T1 - Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation

AU - The Cancer Genome Atlas Research Network

AU - Malta, Tathiane M.

AU - Sokolov, Artem

AU - Gentles, Andrew J.

AU - Burzykowski, Tomasz

AU - Poisson, Laila

AU - Weinstein, John N.

AU - Kamińska, Bożena

AU - Huelsken, Joerg

AU - Omberg, Larsson

AU - Gevaert, Olivier

AU - Colaprico, Antonio

AU - Czerwińska, Patrycja

AU - Mazurek, Sylwia

AU - Mishra, Lopa

AU - Heyn, Holger

AU - Krasnitz, Alex

AU - Godwin, Andrew K.

AU - Lazar, Alexander J.

AU - Caesar-Johnson, Samantha J.

AU - Demchok, John A.

AU - Felau, Ina

AU - Kasapi, Melpomeni

AU - Ferguson, Martin L.

AU - Hutter, Carolyn M.

AU - Sofia, Heidi J.

AU - Tarnuzzer, Roy

AU - Wang, Zhining

AU - Yang, Liming

AU - Zenklusen, Jean C.

AU - Zhang, Jiashan (Julia)

AU - Chudamani, Sudha

AU - Liu, Jia

AU - Lolla, Laxmi

AU - Naresh, Rashi

AU - Pihl, Todd

AU - Sun, Qiang

AU - Wan, Yunhu

AU - Wu, Ye

AU - Cho, Juok

AU - DeFreitas, Timothy

AU - Frazer, Scott

AU - Gehlenborg, Nils

AU - Getz, Gad

AU - Heiman, David I.

AU - Kim, Jaegil

AU - Lawrence, Michael S.

AU - Lin, Pei

AU - Mills, Gordon

AU - Spellman, Paul

AU - Thomas, George

PY - 2018/4/5

Y1 - 2018/4/5

N2 - Cancer progression involves the gradual loss of a differentiated phenotype and acquisition of progenitor and stem-cell-like features. Here, we provide novel stemness indices for assessing the degree of oncogenic dedifferentiation. We used an innovative one-class logistic regression (OCLR) machine-learning algorithm to extract transcriptomic and epigenetic feature sets derived from non-transformed pluripotent stem cells and their differentiated progeny. Using OCLR, we were able to identify previously undiscovered biological mechanisms associated with the dedifferentiated oncogenic state. Analyses of the tumor microenvironment revealed unanticipated correlation of cancer stemness with immune checkpoint expression and infiltrating immune cells. We found that the dedifferentiated oncogenic phenotype was generally most prominent in metastatic tumors. Application of our stemness indices to single-cell data revealed patterns of intra-tumor molecular heterogeneity. Finally, the indices allowed for the identification of novel targets and possible targeted therapies aimed at tumor differentiation. Stemness features extracted from transcriptomic and epigenetic data from TCGA tumors reveal novel biological and clinical insight, as well as potential drug targets for anti-cancer therapies.

AB - Cancer progression involves the gradual loss of a differentiated phenotype and acquisition of progenitor and stem-cell-like features. Here, we provide novel stemness indices for assessing the degree of oncogenic dedifferentiation. We used an innovative one-class logistic regression (OCLR) machine-learning algorithm to extract transcriptomic and epigenetic feature sets derived from non-transformed pluripotent stem cells and their differentiated progeny. Using OCLR, we were able to identify previously undiscovered biological mechanisms associated with the dedifferentiated oncogenic state. Analyses of the tumor microenvironment revealed unanticipated correlation of cancer stemness with immune checkpoint expression and infiltrating immune cells. We found that the dedifferentiated oncogenic phenotype was generally most prominent in metastatic tumors. Application of our stemness indices to single-cell data revealed patterns of intra-tumor molecular heterogeneity. Finally, the indices allowed for the identification of novel targets and possible targeted therapies aimed at tumor differentiation. Stemness features extracted from transcriptomic and epigenetic data from TCGA tumors reveal novel biological and clinical insight, as well as potential drug targets for anti-cancer therapies.

KW - cancer stem cells

KW - dedifferentiation

KW - epigenomic

KW - genomic

KW - machine learning

KW - pan-cancer

KW - stemness

KW - The Cancer Genome Atlas

UR - http://www.scopus.com/inward/record.url?scp=85044967234&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044967234&partnerID=8YFLogxK

U2 - 10.1016/j.cell.2018.03.034

DO - 10.1016/j.cell.2018.03.034

M3 - Article

C2 - 29625051

AN - SCOPUS:85044967234

VL - 173

SP - 338-354.e15

JO - Cell

JF - Cell

SN - 0092-8674

IS - 2

ER -