Back

Systematic Analysis of Human Tissue- and Cell-Specific Metabolic Models Identifies High-Sugar, High Fat Diet Induced Liver Dysregulation

Li, M.; Shi, M.; Zhang, C.; Turkez, H.; Uhlen, M.; Mardinoglu, A.

2026-02-19 bioinformatics
10.64898/2026.02.18.706568 bioRxiv
Show abstract

Human tissues exhibit specialized metabolic functions that are essential for maintaining whole-body metabolic homeostasis. To systematically characterize organ- and cell-type-specific metabolic heterogeneity, we constructed 32 tissue-specific and 81 cell-type-specific enzyme-constrained genome-scale metabolic models (ecGEMs) by integrating the global human metabolic network with the tissue- and single-cell transcriptomic data from the Human Protein Atlas (HPA). Our analysis revealed pronounced differences in metabolic network architecture and activity across the human body, identifying key cell types that drive tissue metabolic functions. To demonstrate the applicability of these models, we employed the liver-specific ecGEM to investigate the metabolic reprogramming induced by a high-sugar, high-fat (HSHF) diet, a primary driver of metabolic dysfunction-associated fatty liver disease (MAFLD). Flux balance analysis revealed a fundamental transition in hepatic metabolism: from a flexible, multi-functional system toward a constrained, lipid-centric regime. This state is characterized by carbohydrate and lipid overload, mitochondrial respiratory dysfunction, and a compromised capacity for reactive oxygen species (ROS) detoxification. These computational predictions were validated through integrative analysis of transcriptomic data from a human MAFLD cohort and metabolomic profiles from an in vivo HSHF rat model. Together, this work provides a comprehensive atlas of human metabolic models, enabling the systematic investigation of metabolic features across tissues and conditions from a systems-level perspective. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=142 SRC="FIGDIR/small/706568v1_ufig1.gif" ALT="Figure 1"> View larger version (40K): org.highwire.dtl.DTLVardef@1015c7org.highwire.dtl.DTLVardef@5c200org.highwire.dtl.DTLVardef@3c982org.highwire.dtl.DTLVardef@a2894d_HPS_FORMAT_FIGEXP M_FIG C_FIG

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.