Home  |   About  |   Theory  |   Tutorials  |   FAQ  |   Bahar Lab  |   Statistics

What is the QuartataWeb Server?
Data on protein-drug and protein-chemical interactions are rapidly accumulating in databases such as DrugBank and STITCH. These data usually reflect observed interactions, while the lack of data for a given protein-drug/chemical pair does not necessarily mean the lack of interaction. Indeed, recent studies, both computational and experimental, highlighted the promiscuity of both proteins and small molecules: many drugs have side effects i.e. they target proteins other than those known in public databases; and many proteins bind chemicals other than those known, opening the way to design repurposable drugs, new chemicals, or polypharmacological treatments. There is a need for efficiently identifying such interactions and disseminating them. The QuartataWeb server is designed to address this need by learning probabilistic latent factor models for protein-drug/chemical interactions and enabling users to efficiently mine known interactions and predict new ones. In addition, QuartataWeb links targets to KEGG pathways and Gene Ontology (GO) terms, thus completing the bridge between drugs/chemicals and pathways.