Computational Health and Life Sciences

What we are doing

Our overarching goal is to build research teams around various themes in life and health sciences. Each assembled team will include data scientists and experts from various coherent domain areas. We aim to develop big-data analytics that have translational values and can be deployed in health care and life sciences. Potential initial themes may include:

• Precision medicine and cardiovascular, diabetes, and metabolic sciences: disease prevention and treatment;
• Systematic mining of electronic health records to predict prognoses and treatment effects for common diseases;
• Integrated multiomics (e.g. genomics, transcriptomics, epigenomics, proteomics, metabolomics, metagenomics) to understand cancer biology and other complex diseases.

We will place emphasis on integrative approaches that connect domain scientists from interrelated areas and experts in data analytics to understand and solve problems from a broader perspective. Instead of addressing each isolated problem, we would strive to address the entire system. As an example, a team for addressing problems arising from the first theme may include experts from food science, nutrition, exercise physiology, microbiome, metabolism, diabetes, cardiovascular medicine, liver cancer, mathematics, statistics, and bioinformatics. To tackle problems of such complexity and range, discussions among team members are indispensable, which can be facilitated by TDAI, with mechanisms and resources to accommodate such activities. We envision that specific themes will grow out from such discussions; research topics and developments from teams will become competitive for large grant opportunities. Training will also be an integrative component of our effort.

Why we are doing it

There has been an explosion of data in all areas of life and health sciences. Big data is now the norm rather than the exception. Such huge sets of data in diverse areas may in fact be inter-connected, and big data analytics, including modeling and data mining methods, are essential for distilling information for convergence into the understanding of an array of phenotypes. Strong bonds between data scientists and domain area experts are needed more than ever. This CoP is conceived to bring the community together to address such challenging problems. Instead of fragmented efforts with limited resources, teams of researchers formed under this CoP can access shared resources facilitated by TDAI, including rich, in-house databases and repositories, to maximize their integrative efforts. This CoP is a platform that can provide researchers speaking different scientific languages a forum to communicate with one another, to exchange ideas, to come up with novel integrative ideas, and to turn such ideas into the development of new technology, new data, or new methodology. The translational values of the research from the teams will address real problems our society faces. The outcomes can have real impacts on life sciences and treatments/preventions of an array of health conditions, including obesity, cancer, and cardiovascular diseases.

Why at Ohio State

Ohio State is uniquely positioned to make fundamental and translational contributions to research in life and health sciences. It has great assets in its faculty, facility, and community. There are also other local research communities highly connected to OSU, including the Research Institute at Nationwide Children’s Hospital. At OSU, there are already many research projects, innovations, and established collaborations leveraging the power of machine learning and other data analytics for solving pressing problems. This CoP will be able to utilize TDAI’s resources to cement existing collaborative relationships and help cultivate new ones, to expand the scope of investigation, and to establish a community well-equipped with the resources to become a leader in synthesizing big data to address problems in life and health sciences.