Biomarker Discovery for Metabolic Dysfunction-associated Steatotic Liver Disease Utilizing Mendelian Randomization, Machine Learning, and External Validation (IMAGE)
Caption
We identify six causal molecular biomarkers (e.g., CNPY4, ENTPD6, HLA-A) and eight clinical biomarkers (e.g., serum total protein) for MASLD across various independent cohorts. Serum total protein levels partially mediated the effect of HLA-A on MASLD, highlighting a novel immune-metabolic pathway. Based on these findings, we develop a random forest model that demonstrates high accuracy in identifying MASLD. Additionally, CNPY4 and ENTPD6 are associated with poor survival in HCC, while low serum total protein levels predicted higher all-cause mortality. These findings support a multi-omics framework for biomarker-driven diagnosis and risk prediction in MASLD.
Credit
Feng Ye
Usage Restrictions
Credit must be given to the creator. Only noncommercial uses of the work are permitted.
License
CC BY-NC