An Evaluation of Cloud Computing for Symptom Science Research: Moving Genomics and Machine Learning Analyses of Cancer Chemotherapy-Related Fatigue to the Cloud

癌症相关性疲劳（CRF）是与癌症及其治疗相关的最常见症状。 中度至重度CRF对患者耐受治疗的能力以及他们的生活质量有负面影响。 生活质量母公司赠款解决了CRF的两个主要知识差距：缺乏风险 预测模型和缺乏对其潜在机制的了解。鉴于这些分析是数据， 资源密集型，他们是不可用的许多症状科学研究人员。在实施方面， 症状科学研究人员的两个主要差距是缺乏必要的计算能力， 资源以及对云部署的好处和成本缺乏了解。症状科学是一种 许多校外和校内护理科学家的突出研究重点。分析的评价 母赠款的管道将确定资源密集型分析，可以有效地部署到 cloud.考虑到云服务的潜在好处， 症状科学研究的云计算以及成本和收益的评估可以指导未来的研究 规划和增加采用云计算在护理研究界。解决这些 限制，我们建议部署和评估RNA-seq和机器学习的性能 管道到云;开发和发布云支持的容器，用于执行表达定量 甲基化（eQTM）映射在云中;并为护理研究提供教育机会 社区介绍我们将这些分析管道部署到云的经验，并提供指导， 协助规划组学和机器学习症状科学研究项目。

Cancer-related fatigue (CRF) is the most common symptom associated with cancer and its treatments. Moderate to severe CRF has a negative impact on patients’ ability to tolerate treatments as well as on their quality of life. The parent grant addressed two of the major knowledge gaps for CRF: the lack of a risk prediction model and a lack of knowledge of its underlying mechanisms. Given these analyses are data and resource intensive, they are unavailable to many symptom science researchers. In terms of implementation, two of the major gaps for symptom science researchers are the lack of access to the necessary computational resources and a lack of understanding of benefits and costs of a cloud deployment. Symptom science is a prominent research focus for many extramural and intramural nurse scientists. An evaluation of the analytic pipelines of the parent grant would identify resource intensive analyses that could be efficiently deployed to the cloud. Given the potential benefits of cloud services, increased knowledge of the opportunities for using the cloud for symptom science research and an evaluation of the costs and benefits could guide future research planning and an increased adoption of cloud computing in the nursing research community. To address these limitations, we propose to deploy and evaluate the performance of the RNA-seq and machine learning pipelines to the cloud; develop and release a cloud-supported container for performing expression quantitative methylation (eQTM) mapping in the cloud; and provide educational opportunities for the nursing research community describing our experience deploying these analytic pipelines to the cloud and providing guidance to aid in planning omics and machine learning symptom science research projects.