Core D: Data Analysis and Research Translation Core

Core D: Data Analysis and Research Translation Core ABSTRACT An overarching theme of this Program Project is to develop novel statistical methods that are understandable and useable by outside investigators in the analysis of their data. In order to accomplish this, two of the major goals of this Program are to demonstrate the application of new methods to real data, and to make our methods accessible to a broad community of researchers. Core D will facilitate both of these goals. First, the Core will generate ‘master’ databases for the benchmarking of new statistical methods as well as publishing novel biological findings. The Core will synthesize essential variables (specific outcomes, predictors, covariates, etc.) from public resources as well as ongoing research studies, and work with Core B to merge data drawn from external data resources into the master databases. Second, the Core will assist and advise on implementing best practices for data analysis with a focus on reproducibility. To ensure reproducibility we will exploit embed technologies such as containers (e.g., Docker), version control software (e.g., Git), and data sharing tools (e.g., Zenodo). Finally, we will develop educational material and run training boot camps and hack fests to teach outside investigators use of the software tools and datasets generated from this program. Such forums will help 1) evaluate the packages, 2) accelerate scientific discoveries, and 3) promote the developed methods to the broader scientific community.