Translational informatics of population health: How large biomolecular and clinical datasets unite

This paper summarizes the workshop content on how the integration of large biomolecular and clinical datasets can enhance the field of population health via translational informatics. Large volumes of data present diverse challenges for existing informatics technology, in terms of computational efficiency, modeling effectiveness, statistical computing, discovery algorithms, and heterogeneous data integration. While accumulating large 'omics measurements on subjects linked with their electronic record remains a challenge, this workshop focuses on non-trivial linkages between large clinical and biomolecular datasets. For example, exposures and clinical datasets can relate through zip codes, while comorbidities and shared molecular mechanisms can relate diseases. Workshop presenters will discuss various methods developed in their respective labs/organizations to overcome the difficulties of combining together such large complex datasets and knowledge to enable the translation to clinical practice for improving health outcomes.