Fig. 3

Distribution of somatic PTEN variants in cancer genomics datasets. A Each nucleotide within the PTEN cDNA was computationally substituted to every other nucleotide to create a list of all possible single-nucleotide variants, and each of these codons was computationally translated to create a list of PTEN missense variants possible through single-nucleotide variation, as well as their relative frequencies of being observed based on degeneracy of the codon table. Approximately 46% of these missense variants had clear abundance and activity classifications (this is the sum of all of the gray lines across the four classes shown in panel A). To aid comparisons for cancer-specific enrichment of PTEN variant classes, these PTEN variant frequencies calculated from mutation only were separated into different phosphatase-abundance groups, and their expected frequencies are shown as thick gray lines. PTEN variants observed in cancer genomics databases were also separated by phosphatase-abundance groups, and their frequencies shown as different colored points. B Loss of activity only variants were next separated into known dominant negatives, as well as other previously uncharacterized loss of activity only variants. The total number of somatic PTEN variants observed in cancer genomics databases with concomitant abundance and activity classifications, is shown as n values next to each cancer type