Machine learning enables new insights into genetic contributions to liver fat accumulation
Mary E. Haas,
James P. Pirruccello,
Samuel N. Friedman,
Minxian Wang,
Connor A. Emdin,
Veeral H. Ajmera,
Tracey G. Simon,
Julian R. Homburger,
Xiuqing Guo,
Matthew Budoff,
Kathleen E. Corey,
Alicia Y. Zhou,
Anthony Philippakis,
Patrick T. Ellinor,
Rohit Loomba,
Puneet Batra,
Amit V. Khera
Affiliations
Mary E. Haas
Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Department of Molecular Biology, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA
James P. Pirruccello
Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Department of Medicine, Harvard Medical School, Boston, MA 02114, USA; Cardiology Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA; Machine Learning for Health, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
Samuel N. Friedman
Machine Learning for Health, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
Minxian Wang
Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
Connor A. Emdin
Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Department of Medicine, Harvard Medical School, Boston, MA 02114, USA
Veeral H. Ajmera
NAFLD Research Center, Department of Medicine, University of California San Diego, La Jolla, CA 92103, USA
Tracey G. Simon
Department of Medicine, Harvard Medical School, Boston, MA 02114, USA; Liver Center, Division of Gastroenterology, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA
Julian R. Homburger
Color Health, Burlingame, CA 94010, USA
Xiuqing Guo
The Lundquist Institute for Biomedical Innovation and Department of Pediatrics, Harbor-UCLA Medical Center, Torrance, CA 90502, USA
Matthew Budoff
The Lundquist Institute for Biomedical Innovation and Department of Pediatrics, Harbor-UCLA Medical Center, Torrance, CA 90502, USA
Kathleen E. Corey
Department of Medicine, Harvard Medical School, Boston, MA 02114, USA; Liver Center, Division of Gastroenterology, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA
Alicia Y. Zhou
Color Health, Burlingame, CA 94010, USA
Anthony Philippakis
Machine Learning for Health, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Eric and Wendy Schmidt Center, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
Patrick T. Ellinor
Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Department of Medicine, Harvard Medical School, Boston, MA 02114, USA; Cardiology Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA; Machine Learning for Health, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
Rohit Loomba
NAFLD Research Center, Department of Medicine, University of California San Diego, La Jolla, CA 92103, USA
Puneet Batra
Machine Learning for Health, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
Amit V. Khera
Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Center for Genomic Medicine, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA; Department of Medicine, Harvard Medical School, Boston, MA 02114, USA; Cardiology Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA; Machine Learning for Health, Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA; Corresponding author
Summary: Excess liver fat, called hepatic steatosis, is a leading risk factor for end-stage liver disease and cardiometabolic diseases but often remains undiagnosed in clinical practice because of the need for direct imaging assessments. We developed an abdominal MRI-based machine-learning algorithm to accurately estimate liver fat (correlation coefficients, 0.97–0.99) from a truth dataset of 4,511 middle-aged UK Biobank participants, enabling quantification in 32,192 additional individuals. 17% of participants had predicted liver fat levels indicative of steatosis, and liver fat could not have been reliably estimated based on clinical factors such as BMI. A genome-wide association study of common genetic variants and liver fat replicated three known associations and identified five newly associated variants in or near the MTARC1, ADH1B, TRIB1, GPAM, and MAST3 genes (p 1.32 per SD score, p < 9 × 10−17). Rare inactivating variants in the APOB or MTTP genes were identified in 0.8% of individuals with steatosis and conferred more than 6-fold risk (p < 2 × 10−5), highlighting a molecular subtype of hepatic steatosis characterized by defective secretion of apolipoprotein B-containing lipoproteins. We demonstrate that our imaging-based machine-learning model accurately estimates liver fat and may be useful in epidemiological and genetic studies of hepatic steatosis.
