© 2016 Nature America, Inc. All rights reserved. NATURE GENETICS ADVANCE ONLINE PUBLICATION 1 LETTERS Intestinal microbiota is known to be important in health and disease. Its composition is influenced by both environmental and host factors. Few large-scale studies have evaluated the association between host genetic variation and the composition of microbiota. We recruited a cohort of 1,561 healthy individuals, of whom 270 belong in 123 families, and found that almost one-third of fecal bacterial taxa were heritable. In addition, we identified 58 SNPs associated with the relative abundance of 33 taxa in 1,098 discovery subjects. Among these, four loci were replicated in a second cohort of 463 subjects: rs62171178 (nearest gene UBR3) associated with Rikenellaceae, rs1394174 (CNTN6) associated with Faecalibacterium, rs59846192 (DMRTB1) associated with Lachnospira, and rs28473221 (SALL3) associated with Eubacterium. After correction for multiple testing, 6 of the 58 associations remained significant, one of which replicated. These results identify associations between specific genetic variants and the gut microbiome. The study cohort comprised 1,561 subjects recruited from 2008 to 2014 as part of the Genetic Environmental Microbial (GEM) Project studying the healthy first-degree relatives of subjects with Crohn’s dis- ease (Table 1, Supplementary Fig. 1a,b, Supplementary Table 1, and Supplementary Note). The discovery cohort (n = 1,098) consisted of asymptomatic self-described white (and genetically consistent with European ancestry) first-degree relatives of subjects with Crohn’s disease. This cohort had a mean age of 20.1 ± 7.8 years (mean ± s.d.; range of 6–35 years), and 54.6% were female. Stool samples were collected for 16S ribosomal DNA sequencing at a minimum depth of 30,000 reads/sample. The three dominant bacte- rial phyla in the fecal samples of the discovery cohort were Firmicutes (relative abundance of 64.4 ± 13.9%), Bacteroidetes (26.7 ± 14.8%), and Actinobacteria (5.0 ± 5.0%) (Supplementary Fig. 2). Of the 127 genera identified, Blautia, Coprococcus, Ruminococcus, Bacteroides, Dorea, Roseburia, Faecalibacterium, Streptococcus, and Oscillospira were found in all subjects. The remaining 118 genera were irregularly observed across subjects, in keeping with the idea of a core microbiome 1 that coexists with highly variable bacterial taxa in the human gut microbiome 2 . A subset of 270 related individuals (mostly siblings) from 123 families in this cohort was used to estimate heritability of the microbiome (Supplementary Fig. 3a–c). Heritability analysis of the microbiota showed that additive genetic factors contributed to abundance for 94 of 249 total bacterial taxa (0.25 < polygenic heritability (H2r) < 0.66, 3.2 × 10 -6 < P < 0.05). Whereas traditional heritability studies report nominal association and confidence intervals, we adjusted P values for correlation of the taxa and multiple testing 3,4 , yielding 20 taxa that remained significant 5,6 (Fig. 1a and Supplementary Table 2). A previous twin study identified 26 heritable taxa 7 . Thirteen of these taxa were nominally replicated in the GEM cohort, of which five remained significant after adjustment for correlation of the taxa and multiple testing (Supplementary Table 2). Seven taxa were not replicated as heritable (0.07 < P < 0.33), and six could not be addressed in our analysis because of absence of the considered taxon in our cohort or because of residual kurtosis >0.8. Our validation of these heritable taxa and identification of additional taxa that are heritable strongly suggest that host genetics is an important factor in the determination of intestinal microbial composition (Supplementary Table 2). Several studies have reported a decrease in bacterial diversity in various diseases such as inflammatory bowel disease (IBD) 8,9 . This raises the possibility that host genotype may influence microbial diversity. However, changes in bacterial diversity may also occur Association of host genome with intestinal microbial composition in a large healthy cohort Williams Turpin 1,2 , Osvaldo Espin-Garcia 3,4 , Wei Xu 4 , Mark S Silverberg 1–3 , David Kevans 1,2 , Michelle I Smith 1,3 , David S Guttman 5,6 , Anne Griffiths 7 , Remo Panaccione 8 , Anthony Otley 9 , Lizhen Xu 4,10 , Konstantin Shestopaloff 4 , Gabriel Moreno-Hagelsieb 11 , GEM Project Research Consortium 12 , Andrew D Paterson 4,10,13 & Kenneth Croitoru 1–3 1 Zane Cohen Centre for Digestive Diseases, Mount Sinai Hospital, Toronto, Ontario, Canada. 2 Division of Gastroenterology, Department of Medicine, University of Toronto, Toronto, Ontario, Canada. 3 Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, Ontario, Canada. 4 Division of Biostatistics, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada. 5 Department of Cell and Systems Biology, University of Toronto, Toronto, Ontario, Canada. 6 Centre for the Analysis of Genome Evolution and Function, University of Toronto, Toronto, Ontario, Canada. 7 Division of Gastroenterology, Hepatology and Nutrition, Department of Pediatrics, The Hospital for Sick Children, Toronto, Ontario, Canada. 8 Inflammatory Bowel Disease Clinic, Division of Gastroenterology and Hepatology, University of Calgary, Calgary, Alberta, Canada. 9 Departement of Pediatrics, IWK Health Centre, Halifax, Nova Scotia, Canada. 10 Genetics and Genome Biology, The Hospital for Sick Children Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada. 11 Department of Biology, Wilfrid Laurier University, Waterloo, Ontario, Canada. 12 A full list of members and affiliations appears at the end of the paper. 13 Division of Epidemiology, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada. Correspondence should be addressed to K.C. ( kcroitoru@mtsinai.on.ca) or A.D.P. (andrew.paterson@sickkids.ca). Received 4 May; accepted 12 September; published online 3 October 2016; doi:10.1038/ng.3693