The computational analysis of tumor cell sensitivity to supertarget deletion.

Abstract

Gene mutations and altered epigenetic regulation of gene expression are characteristic features of malignant neoplasms. Combinations of these abnormalities form molecular features of individual tumors. In the large-scale Dependency Map (DepMap) project, the broad panels of human tumor cell lines are being tested for sensitivity to single gene inactivation. Using DepMap data, we have previously identified a set of genes termed supertargets, the deletion of which significantly reduced the survival of cells of a particular tissue origin while minimally impairing the unrelated cell lines. In the present study, we determined the factors of viability (inhibition of proliferation or death) of cell lines in which the supertarget genes have been deleted. We found that, in 79 % of cases, the reduced survival may be caused by epigenetic changes of gene expression. In the remaining 21 % of cases, it is associated with altered gene structure. Three groups containing different types of gene expression alterations can be distinguished. In the first group, the reduced cell survival correlated with a higher expression of the supertarget gene (e. g., SOX10 and HNF1B). In the second group, a gene different from the deleted supertarget was overexpressed (gene pairs: FOXA1 and SPDEF, TP63 and SERPINB13, etc.). The third group was characterized by correlations between low expression of a certain gene and tumor cell sensitivity (e. g., FAM126A and FAM126B, SMARCA2 and SMARCA4). The genetic changes included GOF mutations (KRAS, BRAF genes, etc.), LOF mutations (STAG1, SMARCA2 genes, etc.), gene fusions (BCR-ABL1, PAX3-FOXO1, etc.), and amplification (CPM, BEST3, etc.). Therefore, many different molecular mechanisms act as predictors of tumor cell response to inhibition of supertarget genes.