Hoggart CJ, Venturini G, Mangino M, Gomez F, Ascari G, Zhao JH, Teumer A, Winkler TW, Tšernikova N, Luan J, Mihailov E, Ehret GB, Zhang W, Lamparter D, Esko T, Macé A, Rüeger S, Bochud PY, Barcella M, Dauvilliers Y, Benyamin B, Evans DM, Hayward C, Lopez MF, Franke L, Russo A, Heid IM, Salvi E, Vendantam S, Arking DE, Boerwinkle E, Chambers JC, Fiorito G, Grallert H, Guarrera S, Homuth G, Huffman JE, Porteous D, Generation Scotland Consortium , LifeLines Cohort study , GIANT Consortium , Moradpour D, Iranzo A, Hebebrand J, Kemp JP, Lammers GJ, Aubert V, Heim MH, Martin NG, Montgomery GW, Peraita-Adrados R, Santamaria J, Negro F, Schmidt CO, Scott RA, Spector TD, Strauch K, Völzke H, Wareham NJ, Yuan W, Bell JT, Chakravarti A, Kooner JS, Peters A, Matullo G, Wallaschofski H, Whitfield JB, Paccaud F, Vollenweider P, Bergmann S, Beckmann JS, Tafti M, Hastie ND, Cusi D, Bochud M, Frayling TM, Metspalu A, Jarvelin MR, Scherag A, Smith GD, Borecki IB, Rousson V, Hirschhorn JN, Rivolta C, Loos RJ, Kutalik Z
PLoS Genet. 10 (7) e1004508 [2014-07-00; online 2014-07-31]
The phenotypic effect of some single nucleotide polymorphisms (SNPs) depends on their parental origin. We present a novel approach to detect parent-of-origin effects (POEs) in genome-wide genotype data of unrelated individuals. The method exploits increased phenotypic variance in the heterozygous genotype group relative to the homozygous groups. We applied the method to >56,000 unrelated individuals to search for POEs influencing body mass index (BMI). Six lead SNPs were carried forward for replication in five family-based studies (of ∼4,000 trios). Two SNPs replicated: the paternal rs2471083-C allele (located near the imprinted KCNK9 gene) and the paternal rs3091869-T allele (located near the SLC2A10 gene) increased BMI equally (beta = 0.11 (SD), P<0.0027) compared to the respective maternal alleles. Real-time PCR experiments of lymphoblastoid cell lines from the CEPH families showed that expression of both genes was dependent on parental origin of the SNPs alleles (P<0.01). Our scheme opens new opportunities to exploit GWAS data of unrelated individuals to identify POEs and demonstrates that they play an important role in adult obesity.