Iman Karimi-Sani - Department of Medical Biotechnology, School of Advanced Medical Sciences and Technologies, Shiraz University of Medical Sciences, Shiraz, Iran
Maryam Fadaei-Dashti - Department of Emergency Medicine, School of Medicine Alborz University of Medical Sciences, Karaj, Iran
Amirabbas Atapour - Department of Medical Biotechnology, School of Advanced Medical Sciences and Technologies, Shiraz University of Medical Sciences, Shiraz, Iran
Mohammad Hossein Morowvat - Department of Pharmaceutical Biotechnology, Shiraz University of Medical Sciences, School of Pharmacy, Shiraz, Iran

Background and aims: Nervous system disorders (NSDs) are often considered the most prevalent,fatal, and most devastating form of nervous system disease. NSDs are indications characterizedby high unmet medical needs, and limited available drugs. NSD drug discovery is a very expensiveprocess that has many unique challenges as a result of which attrition rates and efficiencyare extremely high. The application of artificial intelligence (AI) technologies to the discovery ofNSD drugs has become increasingly attractive. The development of therapeutics for NSDs suchas Alzheimer’s, Parkinsonism, and schizophrenia has been provided with a new direction andthrust from AI technologies.Here, we present a general overview of AI as it relates to drug discovery, as well as an overviewof the recent developments and the applications of AI techniques in NSD drug discovery.Methods: A systematic search of PubMed, Scopus, and Web of Science, Google Scholar wasdone for data till March ۲۰۲۳. Database were searched using the terms ‘artificial intelligence’,‘drug repurposing’, ‘drug repositioning’, ‘nervous system disorders’, ‘machine learning’, ‘deeplearning’, ‘neural network’, ‘random forest’, and ‘support vector machine’. Inclusion criteria forpaper selection were: ۱) Paper must be peer-reviewed. ۲) Journals on which papers are publishedmust be either PubMed, Scopus, or Web of Science indexed. ۳) The paper should use only AItechniques. Exclusion criteria for paper selection were: ۱) MSc and PhD papers. ۲) Duplicatestudies in different databases. Pooled proportions were calculated for categorical variables. Therandom-effects model was used to account for heterogeneity between studies.Results: From the ۸۶۵ identified records [PubMed® (n = ۳۵۹); Scopus (n = ۲۸۵) and or Web ofScience (n = ۲۲۱)], ۹۸ studies were included. NSDs-related articles were classified with subjects:vascular disorders, infections, structural disorders, functional disorders, and degenerations. Moreover,repurposing novel therapeutic candidate drugs for of NSDs was identified. In addition, weprovide a comprehensive background of AI in drug repurposing while specifically focusing on theapplications of a network-based approaches to drug repurposing in NSDs, data sources, and toolsused. Finally, limitations of AI-based approaches in general and specific to a networks are statedalong with future recommendations for better AI-based models.Conclusion: The pharmacological, biological, and epidemiological principles of drug repositioningidentified from the meta-analyses could augment therapeutic development.

Artificial intelligence, network analysis, machine learning, deep learning, drug discovery, drug repositioning, drug re-purposing, nervous system disorders