Discretization of Gene Expression Data Unmasks Molecular Subgroups Recurring in Different Human Cancer Types.
Where this comes from
- Record sourced from PubMed, PMID 27537329.
- Also identified by DOI 10.1371/journal.pone.0161514 and PMC identifier 4990327.
- Licence recorded as CC BY.
- The licence permits redistribution, so the abstract is shown in full and the full text is available from the publisher.
Abstract
Despite the individually different molecular alterations in tumors, the malignancy associated biological traits are strikingly similar. Results of a previous study using renal cell carcinoma (RCC) as a model pointed towards cancer-related features, which could be visualized as three groups by microarray based gene expression analysis. In this study, we used a mathematic model to verify the presence of these groups in RCC as well as in other cancer types. We developed an algorithm for gene-expression deviation profiling for analyzing gene expression data of a total of 8397 patients with 13 different cancer types and normal tissues. We revealed three common Cancer Transcriptomic Profiles (CTPs) which recurred in all investigated tumors. Additionally, CTPs remained robust regardless of the functions or numbers of genes analyzed. CTPs may represent common genetic fingerprints, which potentially reflect the closely related biological traits of human cancers.
Medical subject headings
- Gene Expression Regulation, Neoplastic
- Neoplasms