Rxivist logo

Genome-wide association study of liver fat, iron, and extracellular fluid fraction in the UK Biobank

By Colm O'Dushlaine, Mary Germino, Niek Verweij, Jonas B Nielsen, Ashish Yadav, Christian Benner, Joshua Backman, Nan Li, GHS-REGN DiscovEHR collaboration, Goncalo Abecasis, Aris Baras, Manuel A Ferreira, Luca A Lotta, Johnathon R Walls, Prodromos Parasoglou, Jonathan Marchini

Posted 26 Oct 2021
medRxiv DOI: 10.1101/2021.10.25.21265127

Abdominal magnetic resonance imaging (MRI) represents a non-invasive approach allowing the extraction of clinically informative phenotypes. We developed an automated pipeline to segment liver pixels from abdominal MRI images and apply published models to approximate fat fraction, extracellular fluid fraction and iron content in 40,058 MRIs from the UK Biobank. We then conducted a genome-wide association of these traits using imputed variants (N=37,250 individuals, 11,914,698 variants) and exome sequence data (N=35,274 individuals, 8,287,315 variants). For liver fat we identified 8 novel loci in or near genes MARC1, GCKR, ADH1B, MTTP, TRIB1, GPAM, PNPLA2 and APOH. For liver iron we identified 1 novel locus between the genes ASNSD1 and SLC40A1, an iron transporter involved in hemochromatosis. For extracellular fluid fraction we identified 6 novel loci in or near genes AGMAT, NAT2, MRPL4-S1PR2, FADS1, ABO and HFE, with almost all having prior associations to obesity, liver, iron, or lipid traits.

Download data

  • Downloaded 611 times
  • Download rankings, all-time:
    • Site-wide: 72,213
    • In genetic and genomic medicine: 581
  • Year to date:
    • Site-wide: 17,071
  • Since beginning of last month:
    • Site-wide: 38,119

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide