Schizophrenia: Artificial Intelligence approach and Drug Development

Background and aims: Schizophrenia (SZ) causes psychosis and is associated with considerabledisability and may affect all areas of life functioning. G-protein-coupled receptors (GPCRs), alsoknown as ۷-transmembrane receptors, are the single largest class of drug targets. Consequently,a large number of preclinical assays having GPCRs as molecular targets have been released topublic sources like the Chemical European Molecular Biology Laboratory (ChEMBL) database.One of the aims of this study is to develop a computational model with artificial intelligence (AI)able to predict new GPCRs targeting drugs taking into consideration multiple conditions of theassay. Some receptors are altered in SZ and represent drug targets for antipsychotic therapeuticactivity by using AI.Method: A comprehensive systematic search using the terms such as “Machine Learning”,“Schizophrenia”, “Artificial Intelligence”, “Drug development” and “Drug target” as keywords,was conducted in four Online Databases: PubMed, Web of Science, Embase and Scopus up toJanuary ۲۰۲۳. Also, for screening and data extraction some applications such as “Rayyan” and“Microsoft Excel (۲۰۱۹)” were used. All research that shows SZ drug development and way ofimprovement and with using AI models or machine learning (ML) were included. Reviews andstudies that had not used AI or drug development to improve SZ were excluded. Then studies thatmet our study criteria were critically appraised by two authors independently.Results: We retrieved ۶۸۴ relevant publications from online databases. After a thorough inspectionof abstracts and titles of research and the removal of duplicate publications (n=۳۲), ۴۴۵ studieswere eliminated. In ۷۲ cases of disagreement between two authors, the opinion of the thirdauthor was the determiner. Full texts of ninety-seven papers were reviewed. At last, eleven studiesmet our inclusion criteria and were included in our study. About ۶۴ percent of studies used MLand Support Vector Machine (SVM) algorithms to improve drug development and modeling ofnew drug targets for patients with SZ.Conclusion: Following the results of studies, using AI for the improvement of drug developmentand finding new ways of designing new drugs can be the most effective. AI does this by pinpointingGPCRs coupled receptors that are altered during SZ. In addition, drug development with AIcan be faster and have a lower cost. Despite this, conducting new studies with small sample sizesis one of the limitations of such studies.

کلیدواژه ها:

Schizophrenia (SZ) ، Artificial Intelligence (AI) ، Drug Development ، Drug Target Interaction (DTI)