Fig. 1: Ovarian cancer cell line gene expression and subtyping and copy number signature analysis.

A T-distributed stochastic neighbor embedding (t-SNE) plot of cell lines included in the study. Subtype abbreviations are: DDEC = Dedifferentiated Endometrial Carcinoma, EOC = Endometrioid Carcinoma, HGSOC = High-Grade Serous Ovarian Carcinoma, LGSOC = Low-Grade Serous Ovarian Carcinoma, MUC = Mucinous, OCCC = Ovarian Clear Cell Carcinoma, SCCOHT = Small Cell Carcinoma of the Ovary, Hypercalcemic Type. B Gene ontology terms for top 1000 most variable genes across the 36 cell lines shown as a tree map. Gene ontology terms were reduced by clustering terms with similarity into parent terms. Size of individual gene ontology terms is based on relative proportion of -log10(p_values) weighted across all significant ontology terms plotted. Colors are arbitrary and randomized. C Copy number signature analysis and homologous recombination deficiency (HRD) gene oncoprint for OVCA cell lines in CCLE dataset. The stacked bar plot (top panel) summarizes the weighted contributions of each copy number signature to the overall genomic profile. Copy number signatures are colored by shared etiologies, as described in figure legend to the right. The oncoprint (bottom panel) displays known mutations to HRD genes corresponding to the cell lines in the top panel. IC₅₀ values for cisplatin and carboplatin in our own assays are also shown as a bar graph, as well as calculated HRD score (HRD.sum) and fraction genome altered (fraction_CNV). Percentages on the left summarize the fraction of cell lines carrying an alteration to each gene. Subtype is derived from our literature review.