Multi-Omic Pan-Cancer data from TCGA.

Published: 11 January 2021| Version 3 | DOI: 10.17632/r8p67nfjc8.3
agustin gonzalez reymundez


Data consist of three omic blocks from The Cancer Genome Atlas (TCGA), containing whole-genome profiles of -Gene expression (file GE.RData), -DNA methylation (file METH.RData), and -Copy number variants (file CNV.RData). Omic profiles consist of information from 5,408 tumor samples across 33 cancer types (as matrix rows), and 60,112 features (expression of 20,319 genes, methylation of 28,241 CpG islands, and copy number variant intensity for 11,552 genes). GE profiles by sample corresponded with the logarithm of RNA-Seq counts by gene (Illumina HiSeq RNA V2 platform). METH profiles corresponded with CpG sites B-values from the Illumina HM450 platform, summarized at the CpG island level, using the maximum connectivity approach from the WGCNA R package (Langfelder and Horvath 2008) , and further transformed into M-values (M=beta/(1-beta); Du et al. 2010). Omic blocks were adjusted for batch and tissue specific effects (see Gonzalez-Reymundez and Vazquez (2020) and references therein for further details on quality controls and data edition).



Michigan State University


Cancer, Big Data, Omics