Detecting suicide risk in bipolar disorder patients from lymphoblastoid cell lines genetic signatures

Abstract

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.

Fig. 1. Structural overview of the integrated RNA-seq data analysis and machine learning framework.

a A Flowchart illustrating the process of identifying differentially expressed genes (DEGs) and training and testing the ML models. b A step-by-step training and testing of the ML models.

Fig. 2. Flow chart, Genes and pathways that distinguish the two groups of ‘SUICIDE’ and ‘NON-SUICIDE’.

a Flow chart (b) A heatmap of log10 gene count of the DEGS between the two groups of ‘SUICIDE’ and ‘NON-SUICIDE’. The heat map was organized based on the descending order of the subtraction of the average read count of the two groups (‘SUICIDE’-‘NON-SUICIDE’). c The gene count of the top 30 genes with the lowest p-values is displayed, with red indicating suicidal individuals and blue indicating non-suicidal individuals.

Fig. 3. Dysregulated pathways, disease, and functional annotations in LCLs of BD patients who died by suicide compared patients at a low risk of suicide.

a A diagram displaying the dysregulated KEGG pathways associated with the DEGs between the two groups. b The most prevalent conditions in the DisGeNET database, such as Libman-Sacks Disease, Systemic Lupus Erythematosus, Thrombosis, Neoplasm Invasiveness, Gliosis, Astrocytosis, and various conditions related to pregnancy loss and disorders. c Prominent GAD disease classes, including Cardiovascular, Hematological, Immune, Psychiatric, Renal, and others. d Significant biological processes like Cell Adhesion, Immunity, Chemotaxis, and Inflammatory Response. e Molecular functions associated with the DEGs, including Ion Channels and Growth Factors.

Fig. 4. Top predictive genes for suicide risk.

a A heat map, the heat map was organized based on the descending order of the subtraction of the average read count of the two groups (‘SUICIDE’-‘NON-SUICIDE’). b Gene counts for the most dominant predictor for ML.

Fig. 5. Prediction accuracy by varying the number of genes.

a The variation in model accuracy for a LOR model in relation to the number of genes used as predictor features, ranging from 1–10 genes. b The accuracy of various ML models when specifically using 8 predictors.

Fig. 6. ROC curves and confusion matrices for machine learning models.

a LOR, (b) SVM, (c) RF, (d) NN, and (e) K-NN, along with (f) an overlay of all ROC curves (NN and LOR exhibit overlapping ROC curves).