SFEIES24 Symposia Bone Update (3 abstracts)
Identification of the genetic determinants and clinical implications of bone marrow adiposity using deep learning in the uk biobank
Wei Xu 1 , Ines Mesa-Eguiagaray 1 , David Morris 2,3 , Chengjia Wang 4,3 , Calum Gray 3 , Samuel Sjöström 2 , Giorgos Papanastasiou 5 , Sammy Badr 6 , Julien Paccou 6 , Xue Li 7 , Paul Timmers 8 , Maria Timofeeva 8 , Scott Semple 2,3 , Tom MacGillivray 9 , Evropi Theodoratou 1,10 & William Cawthorn 2
1Centre for Global Health and Molecular Epidemiology, Usher Institute, University of Edinburgh, Edinburgh, United Kingdom; 2University/BHF Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, United Kingdom; 3Edinburgh Imaging, University of Edinburgh, Edinburgh, United Kingdom; 4School of Mathematics and Computer Sciences, Heriot-Watt University, Edinburgh, United Kingdom; 5Archimedes Unit, Athena Research Centre, Marousi, Greece; 6Univ. Lille, CHU Lille, Marrow Adiposity and Bone Laboratory (MABlab) ULR 4490, Department of Rheumatology, Lille, France; 7Department of Big Data in Health Science, School of Public Health and The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China; 8Medical Research Council Human Genetics Unit, Medical Research Council Institute of Genetics & Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom; 9Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom; 10Edinburgh Cancer Research Centre, Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom
Bone marrow adipose tissue (BMAT) is a distinct, major adipose tissue subtype that is a normal feature of mammalian anatomy. However, BMAT’s pathophysiological functions and genetic determinants remain unknown. In humans, bone marrow adiposity is typically measured as the bone marrow fat fraction (BMFF) using magnetic resonance imaging (MRI). Herein, we used deep learning to measure the BMFF of the spine, femoral head, total hip, and femoral diaphysis from MRI of >45,000 participants (>42,000 white, >6,400 non-white) in the UK Biobank imaging study. We then, for the first time, established their heritability and identified the genome- and phenome-wide significant associations for BMFF at each site. Our meta-genome-wide association study (GWAS) in the white population found 67 independent single nucleotide polymorphisms (SNPs) and 54 mapped genes for the femoral head, 147 independent SNPs and 90 mapped genes for the total hip, 134 independent SNPs and 43 mapped genes for the diaphysis, and 174 independent SNPs and 100 mapped genes for the spine. Our multi-ancestry meta-GWAS, including all ethnicities, found 121 independent SNPs and 65 mapped genes for the femoral head, 314 independent SNPs and 98 mapped genes for the total hip, 234 independent SNPs and 63 mapped genes for the diaphysis, and 310 independent SNPs and 121 mapped genes for the spine. These include genes implicated with adipose biology, bone density and/or mesenchymal cell fate, as well as less-expected pathophysiological phenomena. Our phenome-wide association studies (PheWAS) identified 29 diseases associated with BMFF in the femoral head, 36 in the hip, 17 in the diaphysis, and 139 in the spine, collectively spanning over 17 disease categories. As the first GWAS and PheWAS for bone marrow adiposity, these findings provide unprecedented insight into BMAT formation and function, opening new avenues to comprehensively determine the impact of BMAT on human health and disease.
Volume 104
Article tools
My recent searches
My recently viewed abstracts
Authors
Endocrine Abstracts
ISSN 1470-3947 (print) | ISSN 1479-6848 (online)
© Bioscientifica 2024 |
Privacy policy |
Cookie settings
BiosciAbstracts
Bioscientifica Abstracts is the gateway to a series of products that provide a permanent, citable record of abstracts for biomedical and life science conferences.
Find out more