Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells

A goal of cancer research is to reveal cell subsets linked to continuous clinical outcomes to generate new therapeutic and biomarker hypotheses. We introduce a machine learning algorithm, Risk Assessment Population IDentification (RAPID), that is unsupervised and automated, identifies phenotypically...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:eLife Ročník 9
Hlavní autoři: Leelatian, Nalin, Sinnaeve, Justine, Mistry, Akshitkumar M, Barone, Sierra M, Brockman, Asa A, Diggins, Kirsten E, Greenplate, Allison R, Weaver, Kyle D, Thompson, Reid C, Chambless, Lola B, Mobley, Bret C, Ihrie, Rebecca A, Irish, Jonathan M
Médium: Journal Article
Jazyk:angličtina
Vydáno: England eLife Science Publications, Ltd 23.06.2020
eLife Sciences Publications Ltd
eLife Sciences Publications, Ltd
Témata:
Algorithms
Blood cancer
Brain cancer
Brain research
brain tumors
Cancer
Cancer research
Computational and Systems Biology
Cytometry
Data mining
Datasets
Design
Glioblastoma
Glioblastoma - physiopathology
Glioblastomas
Human Biology and Medicine
Humans
Immunohistochemistry
Kinases
Learning algorithms
Machine learning
mass cytomtery
Medical prognosis
Patients
phoshpo-proteins
Pilot Projects
Proteins
Risk assessment
single cell
Statistical analysis
Tumor cells
Tumor Cells, Cultured
Tumors
Unsupervised Machine Learning
Variables
Canada
computational biology
human biology
systems biology
single cell
brain tumors
glioblastoma
medicine
phoshpo-proteins
machine learning
mass cytomtery
human
ISSN:2050-084X, 2050-084X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:A goal of cancer research is to reveal cell subsets linked to continuous clinical outcomes to generate new therapeutic and biomarker hypotheses. We introduce a machine learning algorithm, Risk Assessment Population IDentification (RAPID), that is unsupervised and automated, identifies phenotypically distinct cell populations, and determines whether these populations stratify patient survival. With a pilot mass cytometry dataset of 2 million cells from 28 glioblastomas, RAPID identified tumor cells whose abundance independently and continuously stratified patient survival. Statistical validation within the workflow included repeated runs of stochastic steps and cell subsampling. Biological validation used an orthogonal platform, immunohistochemistry, and a larger cohort of 73 glioblastoma patients to confirm the findings from the pilot cohort. RAPID was also validated to find known risk stratifying cells and features using published data from blood cancer. Thus, RAPID provides an automated, unsupervised approach for finding statistically and biologically significant cells using cytometry data from patient samples.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ObjectType-Undefined-3
These authors contributed equally to this work.
ISSN:2050-084X
2050-084X
DOI:10.7554/eLife.56879