Skip to main content
Fig. 7 | Genome Biology

Fig. 7

From: Tensorial blind source separation for improved analysis of multi-omic data

Fig. 7

Validation of tICA on a multi-omic cancer set. a Manhattan-like plots of IC-35 in gene space, as inferred using tWFOBI on the colon TCGA set, projected along the CNV, DNAm and mRNA axes. Red points highlight genes that had large weights in both CNV and mRNA dimensions (CNV), in both DNAm and mRNA dimensions (DNAm), and the union of these (mRNA). Chromosomes are arranged in increasing order and displayed in alternating colours. b Box plots of the corresponding weights of IC-35 in the sample space, discriminating normal colon (N) from colon cancer (C). P value is from a Wilcoxon rank sum test. c Scatterplots of a driver gene (STX6) between z-score normalised segment level (CNV) and mRNA expression (top panel) and between z-score normalised DNAm level and mRNA expression (lower panel). Colours indicate normal (green) and cancer (red). The regression line, Pearson correlation coefficient and P value are shown. C cancerous, CNV copy-number variation, DNAm DNA methylation, IC independent component, mRNA messenger RNA, N normal, PCC Pearson correlation coefficient, Pos. position

Back to article page