# Exploring the molecular link between obstructive sleep apnea and interstitial cystitis/bladder pain syndrome: A bioinformatics and machine learning study

## TL;DR

This study explores the molecular connection between obstructive sleep apnea and bladder pain syndrome using bioinformatics and machine learning to find shared genes and pathways.

## Contribution

Identifies DUSP9 as a shared biomarker linking OSA and IC/BPS through integrated bioinformatics and machine learning analysis.

## Key findings

- 2,233 and 1,183 differentially expressed genes were identified in OSA and IC/BPS, respectively, with 93 overlapping genes.
- DUSP9 was identified as a key shared gene linking both disorders, forming two-gene signatures for each condition.

## Abstract

Obstructive sleep apnea (OSA) and interstitial cystitis/bladder pain syndrome (IC/BPS) are chronic conditions that significantly impact patients’ quality of life. OSA involves recurrent upper airway obstruction during sleep, causing hypoxia and fragmented sleep linked to cardiovascular and metabolic issues. IC/BPS is defined by chronic pelvic pain and urinary symptoms; its pathophysiology is complex and poorly understood. The overlap in the prevalence of OSA and IC/BPS suggests a possible shared pathophysiological link. This study aimed to identify shared molecular mechanisms and diagnostic biomarkers between OSA and IC/BPS through integrated bioinformatics approaches.

This study used bioinformatics and machine learning to analyze transcriptomic data for OSA and IC/BPS, identifying differential expressed genes (DEGs) and enriched pathways from Gene Expression Omnibus (GEO) database. Weighted gene co-expression network analysis (WGCNA) constructed gene co-expression networks and identified hub genes, while immune infiltration analysis characterized the immune microenvironment. Four machine learning algorithms developed diagnostic models and also identified key markers.

A total of 2,233 DEGs were identified in OSA and 1,183 in IC/BPS, with 93 overlapping genes. Among these, machine learning algorithms identified DUSP9 as the single common gene linking both disorders, forming two-gene signatures for each condition (DUSP9/CCDC68 for OSA and DUSP9/KPNA2 for IC/BPS). Key pathways for OSA included RIG-I-like and NOD-like receptor signaling. In contrast, IC/BPS was linked to cytokine interactions and JAK-STAT signaling. Immune infiltration analysis showed that DUSP9 expression was correlated with CD56dim natural killer cells in OSA and with activated CD4 T cells in IC/BPS, further supporting its role in the immune response associated with these disorders.

This study established DUSP9 as a pivotal shared biomarker and central regulator linking OSA and IC/BPS through integrated bioinformatics analysis.