Abstract
Background: Both lipid metabolism disorders and inflammation are critical contributors to the progression of metabolic-associated fatty liver disease (MAFLD), yet integrated analyses identifying key genes linking them remain scarce. Methods: Differentially expressed genes in MAFLD were extracted from the GSE135251 dataset and intersected with lipid metabolism- and inflammation-related genes from Molecular Signatures Database (MSigDB). Machine learning on GSE135251, followed by validation on GSE89632, identified key genes. Functional enrichment, immune microenvironment profiling, and nomogram analysis were subsequently conducted. Cellular heterogeneity was assessed using the single-cell sequencing (scRNA-seq) dataset GSE186328, and gene expression in MAFLD mice was validated via real-time Polymerase Chain Reaction (PCR). Activators targeting these genes were predicted using Drug Signatures Database (DsigDB). Results: Four genes-FADS1, FADS2, GLB1, and PNPLA3-were identified as key regulators involved in both lipid metabolism disorders and inflammation in MAFLD. These genes were co-enriched in ribosome-related pathways. GLB1 correlated strongly with CD56dim natural killer cells in immune infiltration analysis. A diagnostic nomogram integrating these genes demonstrated exceptional discriminatory power, with Area Under the Curve (AUC) values of 0.98981 for GSE135251 and 0.9204 for GSE89632. ScRNA-seq revealed elevated FADS1, FADS2, and GLB1 expression in MAFLD-associated NK/T cells compared to controls. Real-time PCR confirmed significant upregulation of all four genes in MAFLD mice. Drug prediction identified estradiol as a potential activator targeting these genes. Conclusions: This study identified FADS1, FADS2, GLB1, and PNPLA3 as key genes involved in the progression of MAFLD, linking metabolic dysfunction and inflammation.