Multiple forms of data will be generated from this study. These will include: a) the total number of variants present, b) their presence or absence in the 1000 Genomes Project or ESP5400 non-cancerous genomes, c) their presence or absence in the COSMIC database, d) their SIFT score, e) their POLYPHEN score, f) their chromosomal location, g) the type of each variant (missense, frameshift, nonsense, etc.), h) the overall transcript level for each gene, i) the proportion of expression of each allele (when heterogeneous), and the splicing status. Comparison will be made between the presence of variants and compound activity for the Developmental Therapeutics Program (DTP) 20,602 public compounds and drugs. Versions of this analysis for both single gene variants, as well as combinatorials of variants from a gene will be done. Consideration will be given to which of these variants are projected to have deleterious affects on the protein (as determined by either SIFT or POLYPHEN scores, or the presence of frameshift, premature stop, or read-through variants). A determination will be made as to which variants obtained in this study are also found in clinical cancer samples from The Cancer Genome Atlas (TCGA) repository at http://tcga-data.nci.nih.gov/tcga/. Identification will be made of gene fusions and translocations.
