# Integrative functional genomics reveals transcriptional regulatory function of risk alleles for metabolic liver disease

## TL;DR

This study identifies how non-coding genetic variants contribute to metabolic liver disease by affecting gene regulation and lipid metabolism.

## Contribution

The study reveals cell-type-specific regulatory functions of non-coding risk alleles in metabolic liver disease using integrative functional genomics.

## Key findings

- Non-coding risk variants are enriched in liver chromatin accessible regions bound by cell-specific transcription factors.
- DAVs near lipid metabolism genes modulate gene expression and contribute to liver disease progression.
- Integrative analyses link DAVs to functional target genes and hepatic stellate cell activation.

## Abstract

Genome-wide association studies (GWAS) have identified nearly 100 loci associated with metabolic dysfunction-associated steatotic liver disease (MASLD), but the molecular functions of these variant alleles remain elusive, particularly when they occur in non-coding regions. Here we profiled the chromatin accessibility landscape of liver nuclei from MASLD individuals, and demonstrated these accessible genomic sites were bound by cell type-specific transcription factors (TFs) and enriched for MASLD risk variants, highlighting lineage- and disease state-specific regulation. Using a massively parallel reporter assay (MPRA), we identified hundreds of differential activity variants (DAVs) that operate in a cell type-specific manner or in a stimulus-dependent context by disrupting liver pathogenesis-associated transcriptional regulatory network. Integrative analyses combining liver eQTLs, chromatin looping, and single-cell CRISPRi screening linked these DAVs to functional target genes. Notably, we demonstrated that DAVs located near SLC22A3 and core regulators of triglyceride metabolism (APOA5, ANGPTL3, and LPL) loci modulate their gene expression and contribute to altered lipid metabolism and hepatic stellate cell activation. Furthermore, these DAVs exhibit predictive power in distinguishing MASLD disease risk. Together, these multimodal integration analyses provide insights into the regulatory mechanisms of MASLD progression driven by noncoding genetic risks.