ExScalibur: A High-Performance Cloud-Enabled Suite for Whole Exome Germline and Somatic Mutation Identification
ExScalibur: A High-Performance Cloud-Enabled Suite for Whole Exome Germline and Somatic Mutation Identification
复制标题
DOI：
10.1371/journal.pone.0135800
复制
发表时间：
2015-08-13
期刊：
影响因子：
3.7
通讯作者：
Andrade, Jorge
中科院分区：
文献类型：
--
作者：
Bao, Riyue;Hernandez, Kyle;Andrade, Jorge
全外显子组测序以深度覆盖和低成本促进了与人类疾病相关的致病遗传变异的发现。特别是，从肿瘤/正常对检测体细胞突变提供了对癌症基因组的见解。尽管有大量的公开可用的软件用于检测生殖系和体细胞变异，但变异识别器和比对算法之间的一致性通常有限。成功整合多种方法检测到的变异需要深入了解软件、访问高性能计算资源和高级编程技术。我们提出了Excalibur，一套全自动，高度可扩展和调制的管道，用于整个外显子组数据分析。该套件集成了多种比对和变异识别算法，可准确检测生殖系和体细胞突变，灵敏度和特异性接近99%。Excalibur实现了分析模块的简化执行，实时监控管道进度，强大的错误处理和直观的文档记录，从而提高了结果和工作流程的重现性和共享性。它可以在本地计算机、高性能计算集群和云环境中运行。此外，我们还提供了一个数据分析报告实用程序，以促进结果的可视化，提供质量控制文件的交互式探索，读取比对和变异调用，帮助下游定制潜在的致病突变。Excalibur是开源的，也可以作为亚马逊云上的公共图像。
Whole exome sequencing has facilitated the discovery of causal genetic variants associated with human diseases at deep coverage and low cost. In particular, the detection of somatic mutations from tumor/normal pairs has provided insights into the cancer genome. Although there is an abundance of publicly-available software for the detection of germline and somatic variants, concordance is generally limited among variant callers and alignment algorithms. Successful integration of variants detected by multiple methods requires in-depth knowledge of the software, access to high-performance computing resources, and advanced programming techniques. We present ExScalibur, a set of fully automated, highly scalable and modulated pipelines for whole exome data analysis. The suite integrates multiple alignment and variant calling algorithms for the accurate detection of germline and somatic mutations with close to 99% sensitivity and specificity. ExScalibur implements streamlined execution of analytical modules, real-time monitoring of pipeline progress, robust handling of errors and intuitive documentation that allows for increased reproducibility and sharing of results and workflows. It runs on local computers, high-performance computing clusters and cloud environments. In addition, we provide a data analysis report utility to facilitate visualization of the results that offers interactive exploration of quality control files, read alignment and variant calls, assisting downstream customization of potential disease-causing mutations. ExScalibur is open-source and is also available as a public image on Amazon cloud.