1769-P: Empowering Islet Biology—Advancing Scientific Rigor and Reproducibility, Resource Discoveries, and Effective AI/ML Collaboration via dkNET

The NIDDK Information Network (dkNET; dknet.org) is an open community resource for researchers supported by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), including islet biology investigators. The primary objective is to provide tools, access to diverse resources, and training modules to foster scientific rigor and reproducibility, and collaboration among researchers. dkNET offers various services such as Resource Reports for finding resources and tracking the use and performance of resources (antibodies, model organisms, cell lines, plasmids, biosamples, software tools, and protocols) through the use of Research Resource Identifiers (RRIDs); FAIR Data Resources that provide guidance to meet NIH’s most recent data management and sharing mandates; and the Hypothesis Center that provides tutorials enabling researchers to utilize online datasets and analytics and visualization tools. Building on its success, dkNET is extending its services to provide Artificial Intelligence (AI)/Machine Learning (ML) techniques and cloud computing resources to the NIDDK research community to fully leverage data assets cataloged by dkNET to explore and develop hypotheses. A successful pilot use case, powered by Texera, has been developed to apply machine learning tools to identify the key factors controlling aging of human endocrine cells.