Genetic and Multi‐Omics Insights Into Monocyte Pantothenate‐Mediated Protection in Acute Respiratory Distress Syndrome

Acute respiratory distress syndrome (ARDS) is a severe condition with complex pathogenesis, and emerging evidence highlights the potential role of metabolic factors, though the exact mechanisms are not fully understood. In this study, we used Mendelian randomisation (MR) and multi-omics approaches to investigate the causal relationship between plasma metabolites, immune cell profiles and ARDS risk. MR analysis of 1400 metabolites identified two causal metabolites linked to increased ARDS risk, primarily involved in pantothenate and CoA biosynthesis. Single-cell RNA sequencing of ARDS samples revealed that monocytes exhibited the highest levels of pantothenate synthesis. Intercellular communication and pseudotime analysis suggested that the pantothenate synthesis pathway influenced monocyte differentiation and interactions with other cell types. Gene set enrichment analysis showed that monocytes with high pantothenate synthesis were significantly enriched in phagocytosis-related pathways. Subsequent MR analysis demonstrated that CD33dim HLA DR+ CD11b+%CD33dim HLA DR+ were a risk factor against ARDS. Notably, monocytes with high pantothenate synthesis exhibited decreased expression of antigen presentation markers HLA-DRB5, HLA-DRB1 and HLA-DRA, suggesting that the high pantothenate synthesis monocytes exhibit attenuated antigen presentation and enhanced phagocytic function. Moreover, we developed a diagnostic model using machine learning algorithms. Shapley Additive explanation (SHAP) was leveraged to evaluate the model performance, with CALM2 identified as the most influential feature across the CatBoost and XGBoost models. In summary, this study integrates genetic, multi-omics and machine learning approaches to provide novel insights into the pathogenesis of ARDS and its potential therapeutic strategies targeting monocyte metabolism and function.