Synonymous mutations have been identified to play important roles in cancer development, although they do not modify the protein sequences. However, relatively little research has specifically delineated the functionality of synonymous mutations in cancer. We investigated the nucleotide-based and amino acid-based features of synonymous mutations across 15 cancer types from The Cancer Genome Atlas (TCGA), and revealed novel driver candidates by identifying hotspot mutations. Firstly, synonymous mutations were analyzed between TCGA and 1000 Genomes Project at nucleotide and amino acid levels. We found that C:G → T:A transitions were the most frequent single-base substitutions, and leucine underwent the largest number of synonymous mutations in TCGA due to prevalent C → T transition, which induced the transformation between optimal and non-optimal codons. Next, 97 synonymous hotspot mutations in 86 genes were nominated as candidate drivers with potential cancer risk by considering the mutational rates across different sequence contexts. We observed that non-CpG-island GC transition sequence context was positively selected across most of cancer types, and different sequence contexts under which hotspot mutations occur could be significance for genetic differences and functional features. We also found that the hotspots were more conserved than neutral mutations of hotspot-mutation-containing-genes and frequently happened at leucine. In addition, we mapped hotspots, neutral and non-hotspot mutations of hotspot-mutation-containing-genes to their respective protein domains and found ion transport domain was the most frequent one, which could mediate the cell interaction and had relevant implication for tumor therapy. And the signatures of synonymous hotspots were qualitatively similar with those of harmful missense variants. We illustrated the preferences of cancer associated synonymous mutations, especially hotspots, and laid the groundwork for understanding the synonymous mutations act as drivers in cancer.

中文翻译：

---

15种癌症类型同义突变的突变特征分析

已确定同义突变在癌症发展中起着重要作用，尽管它们不会修饰蛋白质序列。但是，相对较少的研究明确描述了癌症中同义突变的功能。我们研究了来自癌症基因组图谱（TCGA）的15种癌症类型的同义突变的基于核苷酸和氨基酸的特征，并通过识别热点突变揭示了新的候选驱动程序。首先，在核苷酸水平和氨基酸水平上分析了TCGA和100​​0 Genomes Project之间的同义突变。我们发现C：G→T：A转换是最常见的单碱基替换，由于C→T转换很普遍，因此亮氨酸在TCGA中经历了最多数量的同义突变，诱导了最佳和非最佳密码子之间的转换。接下来，通过考虑不同序列背景下的突变率，提名了86个基因中的97个同义热点突变作为具有潜在癌症风险的候选驱动因素。我们观察到，在大多数癌症类型中都积极选择了非CpG-岛GC过渡序列背景，热点突变发生的不同序列背景可能对遗传差异和功能特征具有重要意义。我们还发现，热点比包含热点突变的基因的中性突变更保守，并且亮氨酸经常发生。此外，我们将包含热点突变的基因的热点，中性和非热点突变映射到它们各自的蛋白质域，发现离子转运域是最常见的一个，它可以介导细胞相互作用，对肿瘤治疗具有重要意义。同义热点的特征与有害的错义变体的特征在质量上相似。我们阐明了与癌症相关的同义突变（尤其是热点）的偏好，并为理解同义突变充当癌症驱动因素奠定了基础。