BioGRID: An open resource for biological interactions and network analysis

复杂的物理和遗传相互作用网络决定了所有生物系统的特性

Complex physical and genetic interaction networks determine the properties of all biological systems and underlie human development, health and disease. Decades of biological experiments have identified myriad molecular processes that underpin specific biological processes, described in the primary biomedical literature. More recent technological innovations combined with complete genome sequence information have led to the development of a wide variety high-throughput (HTP) methods to generate physical and genetic interaction data on an unprecedented scale. Because human interaction networks are often directly analogous to networks in more tractable model organisms, it is essential that the hundreds of thousands of biological interactions discovered across the major model organisms, as well as humans, are archived in a well- annotated manner that provides a means for rigorous analysis and computation. To capture, integrate, and interrogate this wealth of data from both the literature and HTP datasets, we developed the BioGRID database as an open repository for physical and genetic interactions (www.thebiogrid.org). BioGRID contains over 2,000,000 total interactions from 75,760 publications. In 2020, BioGRID averaged 151,735 page views, 19,407 unique visitors and 7,537 file downloads per month. Our recently released Open Repository for CRISPR Screens (ORCS), averages 8,725 page views, 1,646 unique visitors, and 268 downloads per month. In addition, the extensive BioGRID data compendium is widely disseminated by many partner databases, meta- databases, and software tools. Here, we propose to markedly enhance the data content, the database architecture, and the user interface of BioGRID. We will expand the amount and types of data available through BioGRID, with a particular focus on translating knowledge from model organism networks to humans using ortholog mapping and a novel framework for mapping phenotypes and diseases across species. We will significantly expand CRISPR-based genetic interactions, chemical and drug interactions, and post-translational modifications, which we will integrate with our core physical and genetic interactions and organize around focused curation efforts around particular biological themes. Use of text-mining algorithms and AI methods will be extended to enhance curation rates and coverage of the proteome. User access to the large datasets in BioGRID will be facilitated by data-rich interfaces, user-defined search and display parameters, and multiple methods of visualization. All software will continue to be open source and engineered toward compatibility and will be complementary with other database and software development efforts. The BioGRID will provide interaction data and software tools to model organism databases and other interested parties without restriction. The BioGRID resource will enable the biomedical research community to access validated biological interaction datasets across model organisms and humans for hypothesis generation and network analysis, and thereby further the general mission of the NIH.