Peter Rose

Dr. Rose is Director of the Structural Bioinformatics Lab and Lead, Bioinformatics and Biomedical Applications at the San Diego Supercomputer Center (SDSC)/UC San Diego. Prior to joining UCSD, he held research and management positions at Pfizer La Jolla, formerly Agouron Pharmaceuticals, where he was instrumental in the establishment of the structure-based drug design platform. He led the RCSB Protein Data Bank group at UCSD and was involved in large-scale data integration with national and international partner sites. He also led an NIH-funded Big Data to Knowledge (BD2K) project to enable large-scale mining and visualization of 3D macromolecular structures. In his current position at SDSC, he is involved in two NSF-funded projects to integrate cross-disciplinary data through knowledge graphs: Project KONQUER aims to integrate biomedical, socio-demographic, and environmental datasets. His current focus is the COVID-19-Net knowledge graph that enables researchers to analyze the interplay of host, pathogen, and the environment. His research interests include structural bioinformatics, structure-based drug design, 3D visualization, and application of big data technologies, knowledge graphs, and machine learning in bioinformatics and biomedicine.

2021 Talk: Integrating Heterogeneous Data Sources into a COVID-19 Knowledge Graph

The COVID-19 pandemic has mobilized researchers worldwide to investigate many aspects of the outbreak, ranging from case statistics, patient demographics, transportation modeling, epidemiological studies, to viral genome sequencing. Relevant data are produced and publically shared at an unprecedented pace and updated daily. Given the urgency of the outbreak and the high levels of velocity and variety of pandemic-related data, efforts have not focused on data interoperability across domains. The avalanche of COVID-19-related data streams from agencies and public and private research teams, with little coordination and without reliance on best interoperability practices, creates enormous challenges for researchers attempting to analyze the pandemic in all its multi-disciplinary complexity and develop a comprehensive policy response. With data collection and analysis efforts largely fragmented and siloed, this goal can be addressed by the roll-out of a comprehensive semantic integration platform that organizes available information into an easily queryable transdisciplinary knowledge system. We developed the COVID-19-Net Knowledge Graph that integrates epidemiological, biological and population characteristic data. The challenges of integrating data across diverse domains, proposed solutions, and calls for action to prepare for future outbreaks will be discussed.