Exploring the Intersection of Schizophrenia, Machine Learning, and Genomics: Scoping Review

Area of Science:

• Psychiatry
• Genomics
• Computational Neuroscience

Background:

• Machine learning (ML) integration with genomic data is growing in psychiatry for complex disorders like schizophrenia.
• Advanced ML techniques offer potential for uncovering multifaceted aspects of schizophrenia.
• A review of ML and genomic data applications enhances understanding of current research and future directions.

Purpose of the Study:

• To conduct a systematic scoping review on the utilization of machine learning algorithms with genomic data in schizophrenia research.

Main Methods:

• A systematic scoping review was performed using MEDLINE, Web of Science, PsycNet (PsycINFO), and Google Scholar (2013-2024).
• Studies focused on the intersection of schizophrenia, genomic data, and machine learning were evaluated.

Main Results:

• 21 studies were thoroughly assessed from an initial 2437 articles.
• Support vector machines were the most common ML algorithm used.
• Genomic data was utilized for predicting schizophrenia, identifying features, drug discovery, classification, and predicting quality of life.

Conclusions:

• Several high-quality studies were identified, but ML applications with genomic data in schizophrenia remain limited.
• Future research is crucial for evaluating model portability and exploring clinical applications.