1 Large Scale Imaging Analytics for In Silico Biomedicine Joel Saltz, Fusheng Wang, George Teodoro, Lee Cooper, Patrick Widener, Jun Kong, David Gutman, Tony Pan, Sharath Cholleti, Ashish Sharma, Daniel Brat, Tahsin Kurc Center for Comprehensive Informatics, the Department of Biomedical Informatics, and the Department of Pathology & Laboratory Medicine Emory University Introduction The ability to quantitatively characterize biological structure and function in detail through in silico experiments 1 has great potential to reveal new insights into disease mechanisms and enable the development of novel preventive approaches and targeted treatments. High-resolution microscopy imaging is playing an increasingly pivotal role in realizing this potential in healthcare delivery and biomedical research. Digital microscopy technology reduces dependence on physical slides; it can also enable more effective ways of screening for disease, classifying disease state, understanding its progression, and evaluating the efficacy of therapeutic strategies. Systematic studies of tumors at the cellular and sub-cellular levels, for example, provide tremendous insight as to how alternations in intercellular signaling occur and allow investigators to study the relationship among morphologic characteristics, cellular-level processes, and genetic, genomic, and protein expression signatures. Studies conducted using tissue slide images and genomic data in the In Silico Brain Tumor Research Center[1] have produced results that reveal morphological subtypes of glioblastoma not previously recognized by pathologists and 1 The term “in silico experiment” broadly refers to an experiment performed on a computer by analyzing, mining, and