Human genetics of metabolic dysfunction–associated steatotic liver disease: from variants to cause to precision treatment

Figures

• 

  Figure 1. PRSs. (A) Sample distribution of risk alleles, which when combined and weighted by effect size, can contribute to calculation of a continuous PRS. (B) Sample PRS plotted versus the percentage of individuals with cirrhosis to show how this score can identify some individuals with high risk of developing the disease.

• 

  Figure 2. Gene-environment interactions. (A) Schematic of gene-environment interactions. In this hypothetical example, the prevalence of hepatic steatosis (y axis) in individuals with low (red) vs. high (blue) environmental risk increases in a dose-dependent manner based on genetic risk (x axis). However, the effect of environmental risk is much greater in those with low genetic risk (absolute difference 10%) versus high genetic risk (absolute difference 30%), indicating a gene-environment interaction. (B) Summary of reported gene-environment interactions for hepatic steatosis severity or liver-related complications in MASLD. The leftmost column lists genes whose variants are known to interact with environmental risk. The top row displays categories of environmental risk factors that interact with genetic risk. Environmental risk factors in red indicate that higher levels of the risk factor confer greater risk of liver disease in those with higher genetic risk, whereas risk factors in blue indicate that higher levels of the risk factor confer lower risk in those with higher genetic risk. Checkmarks show where there is evidence for interactions between specific genes or the polygenic risk…

• 

  Figure 3. Risk gene subgroups associated with PheWAS-identified phenotypes. Panel (A) illustrates the subgroups of risk genes in the context of intracellular and systemic functions linked to their gene products. Panel (B) summarizes the phenotype effects (top row) associated with each risk gene subgroup (leftmost column) that were identified in previously reported PRSs based on human outcomes. Effect sizes for continuous traits are reported as for β values on rank-based inverse normally transformed traits, and as log odds ratio for dichotomous traits. PRSs with significant positive associations are shown as red up arrows, those with significant negative associations are shown as blue down arrows, and those with no significant association (P > 0.05), as hyphens. Effect sizes for continuous traits are reported as β values on rank-based inverse normally transformed traits, and as log(odds ratio) for dichotomous traits. One, two, three, or four arrows indicate absolute value of effect size of <0.04, 0.04-<0.08, 0.08-<0.16, or ≥0.16, respectively. Epidemiologically-expected associations are shown at the bottom and the arrows are agnostic to effect size. VLDL, very low-density lipoproteins;…