OntoQSAR: an ontology for interpreting chemical and biological data in quantitative structure-activity relationship studies

Author(s)

Angelo, Rafaela M. ; Io, Andreia K. ; Almeida, Matheus P. ; Silveira, Rafael G. ; Oliveira, Patricia R. ; Alcazar, Jose J. P. ; Honorio, Kathia M. ; Bettanin, Fernanda ; IEEE

• Document type: Journal article
• Source: 2020 IEEE 14TH INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING (ICSC 2020); v. N/A, p. 4-pg., 2020-01-01.

Abstract

Recent developments in the fields of medicinal chemistry and computer science have made possible to perform rational drug design, helping reduce the cost and time required to discover new therapeutic agents. Particularly, quantitative structure-activity relationship (QSAR) studies have been widely employed to correlate molecular structure to biological activity. Such analyses take into account a set of chemical compounds encoded by numerical descriptors and biological data for generating mathematical models capable to predict biological activity values for structurally similar unknown compounds. Although a huge amount of chemical and biological data has become publicly available, there is an increasing need for integrating and sharing the data from various repositories in order to derive a reusable domain knowledge base. Attempting to overcome such limitation, this article proposes an ontology for QSAR studies, named OntoQSAR, which describes the major concepts in these analyses, like methods used to obtain chemical descriptors and biological properties of chemical compounds. Since the compounds of interest in these studies must be subjected to the same methods for obtaining the necessary numerical descriptors, the developed ontology will allow that information to be reused in other related researches. (AU)

• FAPESP's process: 16/18840-3 - Applying transfer learning techniques and ontologies to QSAR regression models
• Grantee: Patrícia Rufino Oliveira
• Support Opportunities: Research Grants - Research Partnership for Technological Innovation - PITE