Advancements in Data Management and Data Mining Approaches

Health systems are facing a number of challenges in the cost-effective delivery of health care with aging populations and a number of diseases such as obesity, cancer, and diabetes increasing in prevalence. At the same time the life sciences industry is also faced with historically low productivity and a dearth of new drugs to replace medicines reaching loss of exclusivity. Translational medicine has emerged as a science that can help tackle these challenges. The move toward electronic medical records in health systems has provided a rich source of new data for conducting research into the pathophysiology of disease. Increasingly, it is understood that not all drugs work the same in all patients, and tailoring the right drug to the right patient at the right time will help improve medical outcomes while also reducing the cost associated with mistreatment or overtreatment. Key to achieving this is the use of new molecular diagnostic techniques such as next-generation sequencing, which can help scientists and clinicians understand the pathophysiology of disease and also identify which drugs will work in which patients. In this chapter we outline a data management framework that can be used to properly integrate and analyze clinical data from medical records or clinical trials and molecular data from new sequencing technologies. The use of different data integration platforms is discussed and approaches to how these can be used as a backbone to enable data mining. Best practices in data mining are described and common techniques that are used in biomedical research are introduced with some use case examples.