Despite decades of research effort directed towards understanding the basic biology, etiology, prevention, and means of treatment of malignancy, cancer remains one of the most serious health problems for the US population, and is increasingly a global problem with over 14 million new cases of cancer and over 8 million cancer deaths occur every year. Addressing the burden of cancer in the US and worldwide depends on development of a cadre of population-oriented quantitative scientists with skills and knowledge to excel in an increasingly data-rich world. The explosion of biomedical big data has dramatically and irrevocably changed the landscape of cancer research. From molecular investigations studying genomic drivers of cancer to population studies tracking health behaviors and utilization, new data resources are creating unprecedented opportunities to solve the many unanswered questions about cancer. Optimizing the use of these resources and many other complex data?from `omics to administrative databases? will require, in addition to the standard ?tools of the trade? garnered through training in the traditional quantitative disciplines of epidemiology and biostatistics, a deep understanding of different types of data and how they are generated, proficiency in data management and visualization, and knowledge of statistical and machine learning approaches for data analytics. To prepare junior scientists to address the cancer research needs of a data-rich 21st century, we propose to continue our 4 decade old training program for University of Washington (UW) pre-doctoral students and post-doctoral fellows by focusing on ?Developing Data-Driven Cancer Researchers.? Six pre- doctoral and three postdoctoral positions will be filled from established, highly-ranked academic programs in the UW Departments of Epidemiology, Biostatistics, Health Services, Health Metrics as well as interdisciplinary programs such as Nutritional Sciences, Public Health Genetics, and Pharmaceutical Outcomes Research and Policy. Through interdisciplinary didactic and practical experiences, trainees will learn to approach decisions about cancer research strategies from the perspective of the strengths, weaknesses, value, and key analytic features of different types of big data e.g., `omics, clinical, survey, social network, and personal wearable technology-derived health metrics. Features of the training program will include: 1) existing UW courses (such as ?Cancer: Epidemiology and Biology,? ?Biological Basis of Neoplasia,? ?Advanced Health Services Research Methods I: Large Public Databases: Big Data,? ?Machine Learning for Biomedical and Public Health Big Data?); 2) a new, interactive ?The Data of Cancer Research? seminar focused on the generation of cancer data; and 3) completion of a ?Big Data practicum? research exercise that build proficiency in working with cancer data. Thus, the proposed training program will meld extant educational elements with original components unique to the theme of big data to create a unique, value-added learning environment for the next generation of cancer researchers.
This project will give young epidemiologists, biostatisticians, health services researchers, genetic epidemiologists, pharmaceutical outcomes researchers, and health metricians the training they need to access and analyze the wealth of data that are available to address the growing burden of cancer.
Showing the most recent 10 out of 235 publications
