Artificial intelligence and machine learning approaches using gene expression and variant data for personalized...

Area of Science:

• Genomics and Precision Medicine
• Computational Biology
• Artificial Intelligence in Healthcare

Background:

• Precision medicine aims to tailor disease diagnosis and treatment using genetic, environmental, and lifestyle factors.
• Genomic data is central to precision medicine but presents significant analytical challenges.
• Integrating diverse genomic data across populations and diseases requires advanced computational methods.

Purpose of the Study:

• To review and compare artificial intelligence (AI) and machine learning (ML) approaches in genomics and precision medicine.
• To analyze scientific objectives, methodologies, datasets, data sources, ethics, and research gaps.
• To identify widely adapted AI/ML algorithms for predictive diagnostics in genomics.

Main Methods:

• Systematic literature review of high-quality studies published within the last 5 years, indexed in PubMed Central.
• Focused on AI/ML applications in statistical and predictive analyses of whole genome/exome sequencing (gene variants) and RNA-seq/microarrays (gene expression).
• Comparative analysis of identified AI/ML approaches across various genomics studies.

Main Results:

• Identified 32 distinct AI/ML approaches applied in genomics research.
• Reported commonly used AI/ML algorithms for predictive diagnostics across multiple diseases.
• Highlighted the growing role of AI/ML in analyzing complex genomic data for precision medicine applications.

Conclusions:

• AI and ML methods are essential for overcoming the challenges of integrating genomic data into precision medicine.
• The review provides a comprehensive overview of current AI/ML applications, methodologies, and their impact on predictive diagnostics.
• Further research is needed to address identified gaps and optimize AI/ML utilization across diverse populations and diseases.