Identifying stem cell gene expression patterns and phenotypic networks with AutoSOME

Stem cells have the unique property of differentiation and self-renewal and play critical roles in normal development, tissue repair, and disease. To promote systems-wide analysis of cells and tissues, we developed AutoSOME, a machine-learning method for identifying coordinated gene expression patte...

Full description

Saved in:
Bibliographic Details
Published inMethods in molecular biology (Clifton, N.J.) Vol. 1150; p. 115
Main Authors Newman, Aaron M, Cooper, James B
Format Journal Article
LanguageEnglish
Published United States 2014
Subjects
Artificial Intelligence
Biostatistics - methods
Cluster Analysis
Computational Biology - methods
Gene Expression Profiling - methods
Humans
Oligonucleotide Array Sequence Analysis
Phenotype
Sequence Analysis, RNA
Stem Cells - metabolism
Online AccessGet more information

Cover

More Information
Summary:Stem cells have the unique property of differentiation and self-renewal and play critical roles in normal development, tissue repair, and disease. To promote systems-wide analysis of cells and tissues, we developed AutoSOME, a machine-learning method for identifying coordinated gene expression patterns and correlated cellular phenotypes in whole-transcriptome data, without prior knowledge of cluster number or structure. Here, we present a facile primer demonstrating the use of AutoSOME for identification and characterization of stem cell gene expression signatures and for visualization of transcriptome networks using Cytoscape. This protocol should serve as a general foundation for gene expression cluster analysis of stem cells, with applications for studying pluripotency, multi-lineage potential, and neoplastic disease.
ISSN:1940-6029
DOI:10.1007/978-1-4939-0512-6_6