SBIR Phase I: Whole Genome Sequencing Data to Insight in One Hour

这个小企业创新研究（SBIR）项目的更广泛的影响/商业潜力将是在一小时内以四分之一的成本深入了解患者的DNA。这将使医院、诊所和研究中心能够更快地深入研究患者的遗传信息，并将重要的见解反馈给医生，从而更快地做出治疗决定。分析DNA数据有望检测出多种疾病，还可以帮助确定它们的遗传起源，这将是治疗新生儿、罕见疾病患者和孕妇等脆弱病例的关键。通过在一小时内提供完整的DNA数据分析，而不是几天，DNA测试可以成为主流，从而减少患者及其亲属的焦虑。随着需要进行深度DNA分析的患者数量每年翻一番，该项目旨在满足未来大规模计算基因组学的爆炸性需求，实现所有患者的深度DNA分析。SBIR第一阶段项目提议利用最先进的云计算平台的力量，在一小时内提供全基因组测序（WGS）数据的分析。几位关键研究人员已经表明，WGS的数据对于准确洞察和详细分析各种疾病的基础疾病至关重要，包括白血病，乳腺癌，ADHD，阿尔茨海默氏症，先天性心脏病，HIV易感性以及其他非编码区的信息是必需的。然而，WGS数据的计算分析需要几天时间，并且将成为利用关键WGS数据为受影响患者进行个性化治疗的主要瓶颈。该项目旨在云上使用几种高性能计算技术，这些技术将为NGS分析量身定制，并可将该过程加速40倍以上。该项目使用了一种颠覆性的技术，打破了算法在云上的节点上独立工作，该团队创建了一系列软件优化，以提高云资源的利用率。这个优化工具箱正被应用于计算基因组学中常用的软件工具，以实现更快的分析。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) project will be to provide deep insights into the DNA of patients in one hour at one-fourth the cost. This will allow hospitals, clinics and research centers to delve faster into the genetic information of the patients and return essential insights to physicians, leading to faster decisions on therapy. Analyzing DNA data holds the promise of detecting several diseases and can also help in pinpointing their genetic origins, which will be key for treatment of vulnerable cases such as newborn babies, people with rare diseases, and pregnant women. By providing the analysis of whole DNA data in one hour as compared to several days, DNA tests can become mainstream, thereby reducing anxiety among patients and their relatives. As the number of patients for which deep DNA analysis will be required is doubling every year, this project aims to meet the exploding demands of large scale computational genomics of the future and enable deep DNA analysis for all patients.This SBIR Phase I project proposes to use the power of state of the art cloud computing platforms to provide analysis for Whole Genome Sequencing (WGS) data in one hour. Several key researchers have shown that data from WGS is a critical requirement for accurate insights and detailed analysis of underlying diseases for various diseases including leukemia, breast Cancer, ADHD, Alzheimer's, congenital heart disease, HIV susceptibility, as well as others as information in the non-coding region is required. However, the computational analysis for WGS data takes several days and will be the major bottleneck for utilizing key WGS data to personalize the treatment for the affected patient. This project aims to use several high performance computing techniques on the cloud that will be tailored for NGS analyses and can accelerate the process by more than 40 times. This project uses a disruptive technology that breaks algorithms to work independently on nodes on the cloud and the team has created a collection of software optimizations to improve the utilization of cloud resources. This toolbox of optimizations is being applied to commonly used software tools in computational genomics for faster analysis.