P441. Whole-genome sequencing, haplotype characterisation and transcriptomic analysis in an extended multiplex kindred with 40 individuals affected with inflammatory bowel disease


A.P. Levine1, L. Jostins2, G.W. Sewell1, P.J. Smith1, L.B. Lovat3, A.P. Walker1, J.C. Barrett2, A.W. Segal1

1University College London, Division of Medicine, London, United Kingdom; 2Wellcome Trust Sanger Institute, Statistical and Computational Genetics, United Kingdom; 3University College London Hospital, Department of Gastroenterology, London, United Kingdom



Background: The study of large families with many affected individuals is a powerful method for identifying rare disease-causing variants that might contribute to the “missing heritability”. To date, this approach has been of limited value in inflammatory bowel disease (IBD) owing to the moderate size of families described and the polygenic architecture of the disease.

Methods: 4 individuals with IBD belonging to an extended multiplex kindred were identified in the Gastroenterology clinic at UCLH. The family was of Ashkenazi Jewish descent and comprised 800 individuals across 4 generations living in 7 cities on 3 continents. DNA was collected from 160 members including all 40 affecteds. Microarray transcriptomics was performed on cultured monocyte-derived macrophages (n = 8) and histologically non-inflamed rectal biopsies (n = 4). 38 IBD-associated single nucleotide polymorphisms (SNPs) were genotyped in 160 individuals and simulations were conducted to assess the contribution of these to the disease prevalence. 300,000 SNPs were genotyped in 40 affecteds and 60 unaffecteds and a novel parallel method was developed to reconstruct haplotype flow within the family. The exomes (n = 6) and whole-genomes (n = 8) were sequenced.

Results: The affected family members displayed phenotypic heterogeneity with ∼80% having a diagnosis of classical Crohn's disease. All affecteds were non-smokers. The average age at diagnosis was 18 years (95% CI 15–21) and a quarter had undergone surgical resections. Although affecteds had a significant enrichment for IBD-associated SNPs, simulations demonstrated a significant excess of disease cases beyond that expected. Consistent with a complex disease and the presence of a number of sporadic cases, no genome-wide significant linkage was found and there was no gross transcriptomic abnormality shared by all affected individuals studied. High-throughput sequencing identified a number of candidate variants on commonly shared haplotypes and these are currently undergoing bioinformatic prioritisation.

Conclusions: Known IBD-associated SNPs are unable to account for the high degree of familial aggregation observed in this family, the largest IBD pedigree identified to date. Linkage analysis suggests the aetiology of IBD in this family is complex. An integrated approach employing the use of linkage, next-generation sequencing and transcriptomics is being used in ongoing studies to identify rare disease-causing variants and to understand their role in the pathogenesis of IBD in this family.