Fig. 3

RNA-seq analysis of five HGSOC cell lines revealed global transcriptional heterogeneity. a Pearson correlations of gene expression in the HGSOC cell lines OV90, OVCAR3, OVCAR4, OVCAR5, and OVCA432 organized using hierarchal clustering along both rows and columns (n = 4). Scale bar represents coefficient of correlation (r). b Principal component analysis of variance stabilized expression data based on weighted gene lists. X and Y-axes represent the primary and secondary principal components (PC1, PC2), respectively, and report overall contribution to variance in the dataset. Extension to a third PC captured an additional 17% variance but did not improve interpretation of the data. While PC2 separated the two cell lines with the strongest macrophage-inducing potential (OVCAR5 and OVCA432) from the other lines, this analysis did not identify potential candidates due to the sheer number of genes that had positive loadings in PC2. c Number of genes found to be differentially expressed (two-fold up- or down-regulated) in the OVCAR3, OVCAR4, OVCAR5, and OVCA432 cell lines relative to OV90