Related Experiment Video
Updated: Jun 13, 2025
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
Integrative multi-omics investigation of sleep apnea: gut microbiome metabolomics, proteomics and phenome-wide
Shuxu Wei1,2, Ronghuai Shen1, Xiaojia Lu1
1Department of Cardiology, The First Affiliated Hospital of Shantou University Medical College, No.57, Changping Road, Shantou, Guangdong, China.
Background:
Sleep apnea (SA) is linked to various diseases. This study examines the causal link between the gut microbiome and SA, exploring potential predictive factors and target proteins using a multi-omics approach with a Phenome-wide association study (PheWAS).
Methods:
Bidirectional Mendelian Randomization (MR) and Linkage Disequilibrium Score Regression (LDSC) were used to assess the genetic correlation and causal relationships between the gut microbiome and SA. Mediation analysis identified intermediate relationships involving "gut microbiome-inflammatory proteins-SA." Two-sample MR and colocalization analysis in the deCODE and UK Biobank Pharma Proteomics Project (UKB-PPP) databases identified protein quantitative trait loci (pQTL) associated with SA. Validation analysis used Fenland proteins, methylation quantitative trait loci (mQTL), and expression quantitative trait loci (eQTL). PheWAS screened 29 SA-associated SNPs and matched control SNPs (4:1 ratio) from UK Biobank data chosen through MR and LDSC analyses.
Results:
Inverse-variance weighted (IVW) bidirectional MR analysis did not establish a causal link between the gut microbiome and SA. C-C motif chemokine 28 showed causal relationships in both directions (forward IVW, P = 0.0336; reverse IVW, P = 0.0336). Intermediate connections were found between the Holdemanella genus and urinary plasminogen activator levels with SA. TIMP4 protein had a significant causal relationship with SA(IVW method: P > 0.05, PH4 = 96.1%; P = 7.85 × 10-6, PH4 in deCODE = 97.4%). PRIM1 and BMP8 A were identified as potential influencers of SA through mQTL and eQTL analyses. PheWAS suggested body impedance and predicted mass as potential predictors of SA.
Conclusion:
Bidirectional causal relationships exist between SA and inflammatory proteins, with TIMP4 identified as a pathogenic factor and potential therapeutic target. PRIM1 and BMP8 A may impact SA risk. Body impedance and predicted mass predict SA significantly.
