Multi-omics landscape of childhood simple obesity: novel insights into pathogenesis and biomarkers discovery

The increasing incidence of childhood obesity annually has led to a surge in physical and mental health risks, making it a significant global public health concern. This study aimed to discover novel biomarkers of childhood simple obesity through integrative multi-omics analysis, uncovering their potential connections and providing fresh research directions for the complex pathogenesis and treatment strategies.

This study presents the first comprehensive description of the multi-omics characteristics of childhood simple obesity, recognizing promising biomarkers. Immune-related markers offer a new perspective for researching the immunological mechanisms underlying obesity and its associated complications. The revealed interactions among these biomarkers contribute to a deeper understanding the intricate biological regulatory networks associated with obesity.

Transcriptome, untargeted metabolome, and 16 S rDNA sequencing were conducted on subjects to examine transcripts, metabolites in blood, and gut microflora in stool.

Transcriptomic analysis identified 599 differentially expressed genes (DEGs), of which 25 were immune-related genes, and participated in immune pathways such as antimicrobial peptides, neutrophil degranulation, and interferons. The optimal random forest model based on these genes exhibited an AUC of 0.844. The metabolomic analysis examined 71 differentially expressed metabolites (DEMs), including 12 immune-related metabolites. Notably, lauric acid showed an extremely strong positive correlation with BMI and showed a good discriminative power for obesity (AUC = 0.82). DEMs were found to be significantly enriched in four metabolic pathways, namely "Aminoacyl-tRNA biosynthesis", "Valine leucine and isoleucine biosynthesis, and Glycine", "Serine and threonine metabolism", and "Biosynthesis of unsaturated fatty acids". Microbiome analysis revealed 12 differential gut microbiotas (DGMs) at the phylum and genus levels, with p_Firmicutes dominating in the obese group and g_Escherichia-Shigella in the normal group. Subsequently, a Random Forest model was developed based on the DEMs, immune-related DEGs, and metabolites with an AUC value of 0.912. The 14 indicators identified by this model could potentially serve as a set of biomarkers for obesity. The analysis of the inter-omics correlation network found 233 pairs of significant correlations. DEGs BPIFA1, BPI, and SAA1, DEMs Dimethy(tetradecyl)amine, Deoxycholic acid, Pathalic anhydride, and DL-Alanine, and DGMs g_Intestinimonas and g_Turicibacter showed strong connectivity within the network, constituting a large proportion of interactions.