This research aimed to develop a machine learning algorithm to predict suicide risk in bipolar disorder (BD) patients using RNA sequencing analysis of lymphoblastoid cell lines (LCLs). By identifying differentially expressed genes (DEGs) between high and low risk patients and their enrichment in relevant pathways, we gained insights into the molecular mechanisms underlying suicide risk. LCL gene expression analysis revealed pathway enrichment related to primary immunodeficiency, ion channels, and cardiovascular defects. Notably, genes such as LCK, KCNN2, and GRIA1 emerged as pivotal, suggesting their potential roles as biomarkers. Machine learning algorithms trained on a subset of the patients and tested on others demonstrated high accuracy in distinguishing low and high risk of suicide in BD patients. Additionally, the study explored the genetic overlap between suicide-related genes and several psychiatric disorders. Our study enhances the understanding of the complex interplay between genetics and suicidal behaviour, providing a foundation for prevention strategies.

• MeSH
• Receptors, AMPA MeSH
• Biomarkers analysis   MeSH
• Bipolar Disorder * genetics   psychology   MeSH
• Cell Line MeSH
• Child MeSH
• Adult MeSH
• Risk Assessment MeSH
• Humans MeSH
• Adolescent MeSH
• Young Adult MeSH
• Small-Conductance Calcium-Activated Potassium Channels MeSH
• Suicide * psychology   MeSH
• Gene Expression Profiling * MeSH
• Machine Learning * MeSH
• Lymphocyte Specific Protein Tyrosine Kinase p56(lck) MeSH
• Check Tag
• Child MeSH
• Adult MeSH
• Humans MeSH
• Adolescent MeSH
• Young Adult MeSH
• Male MeSH
• Female MeSH
• Publication type
• Journal Article MeSH
• Names of Substances
• Receptors, AMPA MeSH
• Biomarkers MeSH
• glutamate receptor ionotropic, AMPA 1 MeSH Browser
• KCNN2 protein, human MeSH Browser
• LCK protein, human MeSH Browser
• Small-Conductance Calcium-Activated Potassium Channels MeSH
• Lymphocyte Specific Protein Tyrosine Kinase p56(lck) MeSH