Extensive bioinformatics analysis of these datasets of tumor somatic mutations data have revealed the presence of some 500-600 cancer driver genes. The identification of all potential driver mutations affecting cancer genes is essential to implement precision cancer medicine and to understand the interplay of mutation probability and selection in tumor development. Here, we present an in silico saturation mutagenesis approach to identify all drivermutations in 568 cancer genes across 66 tumor types. For most cancer genes the mutation probability across tissues --underpinned by active mutational processes-- influences which driver variants have been observed, although this differs significantly between tumor suppressor and oncogenes. The role of selection is apparent in some of the latter, the observed and unobserved driver mutations of which are equally likely to occur. The number of potential driver mutations in a cancer gene roughly determines how many mutations are available for detection across newly sequenced tumors.

中文翻译：

---

癌症基因的计算机饱和诱变

对肿瘤体细胞突变数据的这些数据集的广泛生物信息学分析表明，存在约500-600个癌症驱动基因。识别影响癌症基因的所有潜在驱动基因突变，对于实施精确的癌症医学以及了解突变概率与肿瘤发展选择之间的相互作用至关重要。在这里，我们提出了一种计算机饱和诱变方法，以识别66种肿瘤类型中的568个癌症基因中的所有驱动突变。对于大多数癌症基因来说，跨组织的突变概率（受活跃的突变过程支持）会影响观察到哪些驱动程序变异，尽管在肿瘤抑制基因和致癌基因之间差异很大。选择的作用在后者中很明显，观察到的和未观察到的驱动程序突变同样可能发生。癌症基因中潜在驱动基因突变的数量大致决定了可用于跨新测序肿瘤检测的突变数量。