Predicting the oncogenicity of missense mutations reported in the International Agency for Cancer Research (IARC) mutation database on p53

Many mutation databases, comprising thousands of reported mutations, are available. Often the clinical significance of the reported mutations is unknown. In this study we developed an algorithm that allows prediction of the clinical significance of missense mutations reported in a mutation database. Nonsense mutations are used as a referent group for this assessment. We used the International Association for Research on Cancer (IARC) mutation database on TP53 to implement the algorithm. First, on the basis of published data [Nachman MW, Crowell SL. 2000. Genetics 156:297-304], we ascribed mutation rates to every single nucleotide substitution (SNS) in the core domain of the TP53 gene. Second, for every possible SNS we computed the expected number of missense mutations, under the assumption that missense mutations are as oncogenic as nonsense ones. The natural logarithm of the ratio of the observed to the expected number of missense mutations (LR) was used as a quantitative measure of oncogenicity (i.e., the ability of a mutation to produce cancer). We estimated the relative oncogenicity of all missense mutations reported in the IARC p53 mutation database, and constructed a profile of oncogenicity of the missense mutations along the DNA-binding region of p53.