Transcription Factor-Centric Approach to Identify Non-Recurring Putative Regulatory Drivers in Cancer.
Transcription Factor-Centric Approach to Identify Non-Recurring Putative Regulatory Drivers in Cancer.
复制标题
DOI：
10.1007/978-3-031-04749-7_3
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
最近对数千个匹配的正常肿瘤样本的基因组进行测序的努力已导致数百万个体细胞突变的鉴定，其中大多数是非编码的。大多数这些突变被认为是过客，但少数非编码突变可能有助于肿瘤的发生或进展，例如肿瘤的发生或进展。导致基因表达失调。识别假定的监管驱动因素的努力主要依赖于有关肿瘤样本中突变复发的信息。然而，在基因组的调控区域中，单个突变很少见于多个供体中。在这里，我们不使用重复信息，而是提出一种根据其对转录因子-DNA 结合的影响程度来识别假定的调控驱动突变的方法。对于每个基因，我们整合了其所有调控区域中突变的影响，并考虑到在感兴趣的队列中在调控 DNA 中观察到的突变谱，我们询问这些影响是否比偶然预期的要大。我们应用我们的方法来分析肝癌数据集中的突变，并提供充足的体细胞突变和基因表达数据。通过结合每个基因所有调控区域的突变影响，我们确定了数十个基因，它们在肿瘤细胞中的调控可能会受到非编码突变的显着干扰。总的来说，我们的结果表明，关注非编码突变的功能影响，而不是它们的复发，有可能识别肿瘤细胞中假定的调控驱动因素和它们失调的基因。
Recent efforts to sequence the genomes of thousands of matched normal-tumor samples have led to the identification of millions of somatic mutations, the majority of which are non-coding. Most of these mutations are believed to be passengers, but a small number of non-coding mutations could contribute to tumor initiation or progression, e.g. by leading to dysregulation of gene expression. Efforts to identify putative regulatory drivers rely primarily on information about the recurrence of mutations across tumor samples. However, in regulatory regions of the genome, individual mutations are rarely seen in more than one donor. Instead of using recurrence information, here we present a method to identify putative regulatory driver mutations based on the magnitude of their effects on transcription factor-DNA binding. For each gene, we integrate the effects of mutations across all its regulatory regions, and we ask whether these effects are larger than expected by chance, given the mutation spectra observed in regulatory DNA in the cohort of interest. We applied our approach to analyze mutations in a liver cancer data set with ample somatic mutation and gene expression data available. By combining the effects of mutations across all regulatory regions of each gene, we identified dozens of genes whose regulation in tumor cells is likely to be significantly perturbed by non-coding mutations. Overall, our results show that focusing on the functional effects of non-coding mutations, rather than their recurrence, has the potential to identify putative regulatory drivers and the genes they dysregulate in tumor cells.