Over the last decade the availability of SNP-trait associations from genome-wide association studies has led to an array of methods for performing Mendelian randomization studies using only summary statistics. A common feature of these methods, besides their intuitive simplicity, is the ability to combine data from several sources, incorporate multiple variants and account for biases due to weak instruments and pleiotropy. With the advent of large and accessible fully-genotyped cohorts such as UK Biobank, there is now increasing interest in understanding how best to apply these well developed summary data methods to individual level data, and to explore the use of more sophisticated causal methods allowing for non-linearity and effect modification. In this paper we describe a general procedure for optimally applying any two sample summary data method using one sample data. Our procedure first performs a meta-analysis of summary data estimates that are intentionally contaminated by collider bias between the genetic instruments and unmeasured confounders, due to conditioning on the observed exposure. These estimates are then used to correct the standard observational association between an exposure and outcome. Simulations are conducted to demonstrate the method's performance against naive applications of two sample summary data MR. We apply the approach to the UK Biobank cohort to investigate the causal role of sleep disturbance on HbA1c levels, an important determinant of diabetes. Our approach can be viewed as a generalization of Dudbridge et al. (Nat. Comm. 10: 1561), who developed a technique to adjust for index event bias when uncovering genetic predictors of disease progression based on case-only data. Our work serves to clarify that in any one sample MR analysis, it can be advantageous to estimate causal relationships by artificially inducing and then correcting for collider bias.

Spousal comparisons have been proposed as a design that can both reduce confounding and estimate effects of the shared adulthood environment. However, assortative mating, the process by which individuals select phenotypically (dis)similar mates, could distort associations when comparing spouses. We evaluated the use of spousal comparisons, as in the within-spouse pair (WSP) model, for aetiological research such as genetic association studies. We demonstrated that the WSP model can reduce confounding but may be susceptible to collider bias arising from conditioning on assorted spouse pairs. Analyses using UK Biobank spouse pairs found that WSP genetic association estimates were smaller than estimates from random pairs for height, educational attainment, and BMI variants. Within-sibling pair estimates, robust to demographic and parental effects, were also smaller than random pair estimates for height and educational attainment, but not for BMI. WSP models, like other within-family models, may reduce confounding from demographic factors in genetic association estimates, and so could be useful for triangulating evidence across study designs to assess the robustness of findings. However, WSP estimates should be interpreted with caution due to potential collider bias.

The Icelandic population has been sampled in many disease association studies, providing a strong motivation to understand the structure of this population and its ramifications for disease gene mapping. Previous work using 40 microsatellites showed that the Icelandic population is relatively homogeneous, but exhibits subtle population structure that can bias disease association statistics. Here, we show that regional geographic ancestries of individuals from Iceland can be distinguished using 292,289 autosomal single-nucleotide polymorphisms (SNPs). We further show that subpopulation differences are due to genetic drift since the settlement of Iceland 1100 years ago, and not to varying contributions from different ancestral populations. A consequence of the recent origin of Icelandic population structure is that allele frequency differences follow a null distribution devoid of outliers, so that the risk of false positive associations due to stratification is minimal. Our results highlight an important distinction between population differences attributable to recent drift and those arising from more ancient divergence, which has implications both for association studies and for efforts to detect natural selection using population differentiation.

The detrimental health effects of smoking are well-known, but the impact of regular nicotine use without exposure to the other constituents of tobacco is less clear. Given the increasing daily use of alternative nicotine delivery systems, such as e-cigarettes, it is increasingly important to understand and separate the effects of nicotine use from the impact of tobacco smoke exposure. Using a multivariable Mendelian randomisation framework, we explored the direct effects of nicotine compared with the non-nicotine constituents of tobacco smoke on health outcomes (lung cancer, chronic obstructive pulmonary disease [COPD], forced expiratory volume in one second [FEV-1], forced vital capacity [FVC], coronary heart disease [CHD], and heart rate [HR]). We used Genome-Wide Association Study (GWAS) summary statistics from Buchwald and colleagues, the GWAS and Sequencing Consortium of Alcohol and Nicotine, the International Lung Cancer Consortium, and UK Biobank. Increased nicotine metabolism increased the risk of COPD, lung cancer, and lung function in the univariable analysis. However, when accounting for smoking heaviness in the multivariable analysis, we found that increased nicotine metabolite ratio (indicative of decreased nicotine exposure per cigarette smoked) decreases heart rate (b = -0.30, 95% CI -0.50 to -0.10) and lung function (b = -33.33, 95% CI -41.76 to -24.90). There was no clear evidence of an effect on the remaining outcomes. The results suggest that these smoking-related outcomes are not due to nicotine exposure but are caused by the other components of tobacco smoke; however, there are multiple potential sources of bias, and the results should be triangulated using evidence from a range of methodologies.

Alu elements are trans-mobilized by the autonomous non-LTR retroelement, LINE-1 (L1). Alu-induced insertion mutagenesis contributes to about 0.1% human genetic disease and is responsible for the majority of the documented instances of human retroelement insertion-induced disease. Here we introduce a SINE recovery method that provides a complementary approach for comprehensive analysis of the impact and biological mechanisms of Alu retrotransposition. Using this approach, we recovered 226 de novo tagged Alu inserts in HeLa cells. Our analysis reveals that in human cells marked Alu inserts driven by either exogenously supplied full length L1 or ORF2 protein are indistinguishable. Four percent of de novo Alu inserts were associated with genomic deletions and rearrangements and lacked the hallmarks of retrotransposition. In contrast to L1 inserts, 5' truncations of Alu inserts are rare, as most of the recovered inserts (96.5%) are full length. De novo Alus show a random pattern of insertion across chromosomes, but further characterization revealed an Alu insertion bias exists favoring insertion near other SINEs, highly conserved elements, with almost 60% landing within genes. De novo Alu inserts show no evidence of RNA editing. Priming for reverse transcription rarely occurred within the first 20 bp (most 5') of the A-tail. The A-tails of recovered inserts show significant expansion, with many at least doubling in length. Sequence manipulation of the construct led to the demonstration that the A-tail expansion likely occurs during insertion due to slippage by the L1 ORF2 protein. We postulate that the A-tail expansion directly impacts Alu evolution by reintroducing new active source elements to counteract the natural loss of active Alus and minimizing Alu extinction.

Genome-wide association studies (GWAS) are commonly used to identify genomic variants that are associated with complex traits, and estimate the magnitude of this association for each variant. However, it has been widely observed that the association estimates of variants tend to be lower in a replication study than in the study that discovered those associations. A phenomenon known as Winner's Curse is responsible for this upward bias present in association estimates of significant variants in the discovery study. We review existing Winner's Curse correction methods which require only GWAS summary statistics in order to make adjustments. In addition, we propose modifications to improve existing methods and propose a novel approach which uses the parametric bootstrap. We evaluate and compare methods, first using a wide variety of simulated data sets and then, using real data sets for three different traits. The metric, estimated mean squared error (MSE) over significant SNPs, was primarily used for method assessment. Our results indicate that widely used conditional likelihood based methods tend to perform poorly. The other considered methods behave much more similarly, with our proposed bootstrap method demonstrating very competitive performance. To complement this review, we have developed an R package, 'winnerscurse' which can be used to implement these various Winner's Curse adjustment methods to GWAS summary statistics.

Genome-wide association studies (GWAS) have uncovered numerous genetic variants (SNPs) that are associated with blood pressure (BP). Genetic variants may lead to BP changes by acting on intermediate molecular phenotypes such as coded protein sequence or gene expression, which in turn affect BP variability. Therefore, characterizing genes whose expression is associated with BP may reveal cellular processes involved in BP regulation and uncover how transcripts mediate genetic and environmental effects on BP variability. A meta-analysis of results from six studies of global gene expression profiles of BP and hypertension in whole blood was performed in 7017 individuals who were not receiving antihypertensive drug treatment. We identified 34 genes that were differentially expressed in relation to BP (Bonferroni-corrected p<0.05). Among these genes, FOS and PTGS2 have been previously reported to be involved in BP-related processes; the others are novel. The top BP signature genes in aggregate explain 5%-9% of inter-individual variance in BP. Of note, rs3184504 in SH2B3, which was also reported in GWAS to be associated with BP, was found to be a trans regulator of the expression of 6 of the transcripts we found to be associated with BP (FOS, MYADM, PP1R15A, TAGAP, S100A10, and FGBP2). Gene set enrichment analysis suggested that the BP-related global gene expression changes include genes involved in inflammatory response and apoptosis pathways. Our study provides new insights into molecular mechanisms underlying BP regulation, and suggests novel transcriptomic markers for the treatment and prevention of hypertension.

Non-linear Mendelian randomization is an extension to standard Mendelian randomization to explore the shape of the causal relationship between an exposure and outcome using an instrumental variable. A stratification approach to non-linear Mendelian randomization divides the population into strata and calculates separate instrumental variable estimates in each stratum. However, the standard implementation of stratification, referred to as the residual method, relies on strong parametric assumptions of linearity and homogeneity between the instrument and the exposure to form the strata. If these stratification assumptions are violated, the instrumental variable assumptions may be violated in the strata even if they are satisfied in the population, resulting in misleading estimates. We propose a new stratification method, referred to as the doubly-ranked method, that does not require strict parametric assumptions to create strata with different average levels of the exposure such that the instrumental variable assumptions are satisfied within the strata. Our simulation study indicates that the doubly-ranked method can obtain unbiased stratum-specific estimates and appropriate coverage rates even when the effect of the instrument on the exposure is non-linear or heterogeneous. Moreover, it can also provide unbiased estimates when the exposure is coarsened (that is, rounded, binned into categories, or truncated), a scenario that is common in applied practice and leads to substantial bias in the residual method. We applied the proposed doubly-ranked method to investigate the effect of alcohol intake on systolic blood pressure, and found evidence of a positive effect of alcohol intake, particularly at higher levels of alcohol consumption.

Head and neck squamous cell carcinoma (HNSCC), which includes cancers of the oral cavity and oropharynx, is a cause of substantial global morbidity and mortality. Strategies to reduce disease burden include discovery of novel therapies and repurposing of existing drugs. Statins are commonly prescribed for lowering circulating cholesterol by inhibiting HMG-CoA reductase (HMGCR). Results from some observational studies suggest that statin use may reduce HNSCC risk. We appraised the relationship of genetically-proxied cholesterol-lowering drug targets and other circulating lipid traits with oral (OC) and oropharyngeal (OPC) cancer risk using two-sample Mendelian randomization (MR). For the primary analysis, germline genetic variants in HMGCR, NPC1L1, CETP, PCSK9 and LDLR were used to proxy the effect of low-density lipoprotein cholesterol (LDL-C) lowering therapies. In secondary analyses, variants were used to proxy circulating levels of other lipid traits in a genome-wide association study (GWAS) meta-analysis of 188,578 individuals. Both primary and secondary analyses aimed to estimate the downstream causal effect of cholesterol lowering therapies on OC and OPC risk. The second sample for MR was taken from a GWAS of 6,034 OC and OPC cases and 6,585 controls (GAME-ON). Analyses were replicated in UK Biobank, using 839 OC and OPC cases and 372,016 controls and the results of the GAME-ON and UK Biobank analyses combined in a fixed-effects meta-analysis. We found limited evidence of a causal effect of genetically-proxied LDL-C lowering using HMGCR, NPC1L1, CETP or other circulating lipid traits on either OC or OPC risk. Genetically-proxied PCSK9 inhibition equivalent to a 1 mmol/L (38.7 mg/dL) reduction in LDL-C was associated with an increased risk of OC and OPC combined (OR 1.8 95%CI 1.2, 2.8, p = 9.31 x10-05), with good concordance between GAME-ON and UK Biobank (I2 = 22%). Effects for PCSK9 appeared stronger in relation to OPC (OR 2.6 95%CI 1.4, 4.9) than OC (OR 1.4 95%CI 0.8, 2.4). LDLR variants, resulting in genetically-proxied reduction in LDL-C equivalent to a 1 mmol/L (38.7 mg/dL), reduced the risk of OC and OPC combined (OR 0.7, 95%CI 0.5, 1.0, p = 0.006). A series of pleiotropy-robust and outlier detection methods showed that pleiotropy did not bias our findings. We found limited evidence for a role of cholesterol-lowering in OC and OPC risk, suggesting previous observational results may have been confounded. There was some evidence that genetically-proxied inhibition of PCSK9 increased risk, while lipid-lowering variants in LDLR, reduced risk of combined OC and OPC. This result suggests that the mechanisms of action of PCSK9 on OC and OPC risk may be independent of its cholesterol lowering effects; however, this was not supported uniformly across all sensitivity analyses and further replication of this finding is required.

Chemerin is an adipokine proposed to link obesity and chronic inflammation of adipose tissue. Genetic factors determining chemerin release from adipose tissue are yet unknown. We conducted a meta-analysis of genome-wide association studies (GWAS) for serum chemerin in three independent cohorts from Europe: Sorbs and KORA from Germany and PPP-Botnia from Finland (total N = 2,791). In addition, we measured mRNA expression of genes within the associated loci in peripheral mononuclear cells by micro-arrays, and within adipose tissue by quantitative RT-PCR and performed mRNA expression quantitative trait and expression-chemerin association studies to functionally substantiate our loci. Heritability estimate of circulating chemerin levels was 16.2% in the Sorbs cohort. Thirty single nucleotide polymorphisms (SNPs) at chromosome 7 within the retinoic acid receptor responder 2 (RARRES2)/Leucine Rich Repeat Containing (LRRC61) locus reached genome-wide significance (p<5.0×10-8) in the meta-analysis (the strongest evidence for association at rs7806429 with p = 7.8×10-14, beta = -0.067, explained variance 2.0%). All other SNPs within the cluster were in linkage disequilibrium with rs7806429 (minimum r2 = 0.43 in the Sorbs cohort). The results of the subgroup analyses of males and females were consistent with the results found in the total cohort. No significant SNP-sex interaction was observed. rs7806429 was associated with mRNA expression of RARRES2 in visceral adipose tissue in women (p<0.05 after adjusting for age and body mass index). In conclusion, the present meta-GWAS combined with mRNA expression studies highlights the role of genetic variation in the RARRES2 locus in the regulation of circulating chemerin concentrations.

Recent work has shown that much of the missing heritability of complex traits can be resolved by estimates of heritability explained by all genotyped SNPs. However, it is currently unknown how much heritability is missing due to poor tagging or additional causal variants at known GWAS loci. Here, we use variance components to quantify the heritability explained by all SNPs at known GWAS loci in nine diseases from WTCCC1 and WTCCC2. After accounting for expectation, we observed all SNPs at known GWAS loci to explain 1.29 x more heritability than GWAS-associated SNPs on average (P=3.3 x 10⁻⁵). For some diseases, this increase was individually significant: 2.07 x for Multiple Sclerosis (MS) (P=6.5 x 10⁻⁹) and 1.48 x for Crohn's Disease (CD) (P = 1.3 x 10⁻³); all analyses of autoimmune diseases excluded the well-studied MHC region. Additionally, we found that GWAS loci from other related traits also explained significant heritability. The union of all autoimmune disease loci explained 7.15 x more MS heritability than known MS SNPs (P < 1.0 x 10⁻¹⁶ and 2.20 x more CD heritability than known CD SNPs (P = 6.1 x 10⁻⁹), with an analogous increase for all autoimmune diseases analyzed. We also observed significant increases in an analysis of > 20,000 Rheumatoid Arthritis (RA) samples typed on ImmunoChip, with 2.37 x more heritability from all SNPs at GWAS loci (P = 2.3 x 10⁻⁶) and 5.33 x more heritability from all autoimmune disease loci (P < 1 x 10⁻¹⁶ compared to known RA SNPs (including those identified in this cohort). Our methods adjust for LD between SNPs, which can bias standard estimates of heritability from SNPs even if all causal variants are typed. By comparing adjusted estimates, we hypothesize that the genome-wide distribution of causal variants is enriched for low-frequency alleles, but that causal variants at known GWAS loci are skewed towards common alleles. These findings have important ramifications for fine-mapping study design and our understanding of complex disease architecture.

Reassortments and point mutations are two major contributors to diversity of Influenza A virus; however, the link between these two processes is unclear. It has been suggested that reassortments provoke a temporary increase in the rate of amino acid changes as the viral proteins adapt to new genetic environment, but this phenomenon has not been studied systematically. Here, we use a phylogenetic approach to infer the reassortment events between the 8 segments of influenza A H3N2 virus since its emergence in humans in 1968. We then study the amino acid replacements that occurred in genes encoded in each segment subsequent to reassortments. In five out of eight genes (NA, M1, HA, PB1 and NS1), the reassortment events led to a transient increase in the rate of amino acid replacements on the descendant phylogenetic branches. In NA and HA, the replacements following reassortments were enriched with parallel and/or reversing replacements; in contrast, the replacements at sites responsible for differences between antigenic clusters (in HA) and at sites under positive selection (in NA) were underrepresented among them. Post-reassortment adaptive walks contribute to adaptive evolution in Influenza A: in NA, an average reassortment event causes at least 2.1 amino acid replacements in a reassorted gene, with, on average, 0.43 amino acid replacements per evolving post-reassortment lineage; and at least ~9% of all amino acid replacements are provoked by reassortments.

The surface proteins hemagglutinin (HA) and neuraminidase (NA) of human influenza A virus evolve under selection pressures to escape adaptive immune responses and antiviral drug treatments. In addition to these external selection pressures, some mutations in HA are known to affect the adaptive landscape of NA, and vice versa, because these two proteins are physiologically interlinked. However, the extent to which evolution of one protein affects the evolution of the other one is unknown. Here we develop a novel phylogenetic method for detecting the signatures of such genetic interactions between mutations in different genes - that is, inter-gene epistasis. Using this method, we show that influenza surface proteins evolve in a coordinated way, with mutations in HA affecting subsequent spread of mutations in NA and vice versa, at many sites. Of particular interest is our finding that the oseltamivir-resistance mutations in NA in subtype H1N1 were likely facilitated by prior mutations in HA. Our results illustrate that the adaptive landscape of a viral protein is remarkably sensitive to its genomic context and, more generally, that the evolution of any single protein must be understood within the context of the entire evolving genome.

Third-generation cephalosporins are a class of β-lactam antibiotics that are often used for the treatment of human infections caused by Gram-negative bacteria, especially Escherichia coli. Worryingly, the incidence of human infections caused by third-generation cephalosporin-resistant E. coli is increasing worldwide. Recent studies have suggested that these E. coli strains, and their antibiotic resistance genes, can spread from food-producing animals, via the food-chain, to humans. However, these studies used traditional typing methods, which may not have provided sufficient resolution to reliably assess the relatedness of these strains. We therefore used whole-genome sequencing (WGS) to study the relatedness of cephalosporin-resistant E. coli from humans, chicken meat, poultry and pigs. One strain collection included pairs of human and poultry-associated strains that had previously been considered to be identical based on Multi-Locus Sequence Typing, plasmid typing and antibiotic resistance gene sequencing. The second collection included isolates from farmers and their pigs. WGS analysis revealed considerable heterogeneity between human and poultry-associated isolates. The most closely related pairs of strains from both sources carried 1263 Single-Nucleotide Polymorphisms (SNPs) per Mbp core genome. In contrast, epidemiologically linked strains from humans and pigs differed by only 1.8 SNPs per Mbp core genome. WGS-based plasmid reconstructions revealed three distinct plasmid lineages (IncI1- and IncK-type) that carried cephalosporin resistance genes of the Extended-Spectrum Beta-Lactamase (ESBL)- and AmpC-types. The plasmid backbones within each lineage were virtually identical and were shared by genetically unrelated human and animal isolates. Plasmid reconstructions from short-read sequencing data were validated by long-read DNA sequencing for two strains. Our findings failed to demonstrate evidence for recent clonal transmission of cephalosporin-resistant E. coli strains from poultry to humans, as has been suggested based on traditional, low-resolution typing methods. Instead, our data suggest that cephalosporin resistance genes are mainly disseminated in animals and humans via distinct plasmids.

Meta-analysis of genetic association studies increases sample size and the power for mapping complex traits. Existing methods are mostly developed for datasets without missing values, i.e. the summary association statistics are measured for all variants in contributing studies. In practice, genotype imputation is not always effective. This may be the case when targeted genotyping/sequencing assays are used or when the un-typed genetic variant is rare. Therefore, contributed summary statistics often contain missing values. Existing methods for imputing missing summary association statistics and using imputed values in meta-analysis, approximate conditional analysis, or simple strategies such as complete case analysis all have theoretical limitations. Applying these approaches can bias genetic effect estimates and lead to seriously inflated type-I or type-II errors in conditional analysis, which is a critical tool for identifying independently associated variants. To address this challenge and complement imputation methods, we developed a method to combine summary statistics across participating studies and consistently estimate joint effects, even when the contributed summary statistics contain large amounts of missing values. Based on this estimator, we proposed a score statistic called PCBS (partial correlation based score statistic) for conditional analysis of single-variant and gene-level associations. Through extensive analysis of simulated and real data, we showed that the new method produces well-calibrated type-I errors and is substantially more powerful than existing approaches. We applied the proposed approach to one of the largest meta-analyses to date for the cigarettes-per-day phenotype. Using the new method, we identified multiple novel independently associated variants at known loci for tobacco use, which were otherwise missed by alternative methods. Together, the phenotypic variance explained by these variants was 1.1%, improving that of previously reported associations by 71%. These findings illustrate the extent of locus allelic heterogeneity and can help pinpoint causal variants.

The epidemiologically most important mechanism of antibiotic resistance in Staphylococcus aureus is associated with mecA-an acquired gene encoding an extra penicillin-binding protein (PBP2a) with low affinity to virtually all β-lactams. The introduction of mecA into the S. aureus chromosome has led to the emergence of methicillin-resistant S. aureus (MRSA) pandemics, responsible for high rates of mortality worldwide. Nonetheless, little is known regarding the origin and evolution of mecA. Different mecA homologues have been identified in species belonging to the Staphylococcus sciuri group representing the most primitive staphylococci. In this study we aimed to identify evolutionary steps linking these mecA precursors to the β-lactam resistance gene mecA and the resistance phenotype. We sequenced genomes of 106 S. sciuri, S. vitulinus and S. fleurettii strains and determined their oxacillin susceptibility profiles. Single-nucleotide polymorphism (SNP) analysis of the core genome was performed to assess the genetic relatedness of the isolates. Phylogenetic analysis of the mecA gene homologues and promoters was achieved through nucleotide/amino acid sequence alignments and mutation rates were estimated using a Bayesian analysis. Furthermore, the predicted structure of mecA homologue-encoded PBPs of oxacillin-susceptible and -resistant strains were compared. We showed for the first time that oxacillin resistance in the S. sciuri group has emerged multiple times and by a variety of different mechanisms. Development of resistance occurred through several steps including structural diversification of the non-binding domain of native PBPs; changes in the promoters of mecA homologues; acquisition of SCCmec and adaptation of the bacterial genetic background. Moreover, our results suggest that it was exposure to β-lactams in human-created environments that has driven evolution of native PBPs towards a resistance determinant. The evolution of β-lactam resistance in staphylococci highlights the numerous resources available to bacteria to adapt to the selective pressure of antibiotics.

More than 800 published genetic association studies have implicated dozens of potential risk loci in Parkinson's disease (PD). To facilitate the interpretation of these findings, we have created a dedicated online resource, PDGene, that comprehensively collects and meta-analyzes all published studies in the field. A systematic literature screen of -27,000 articles yielded 828 eligible articles from which relevant data were extracted. In addition, individual-level data from three publicly available genome-wide association studies (GWAS) were obtained and subjected to genotype imputation and analysis. Overall, we performed meta-analyses on more than seven million polymorphisms originating either from GWAS datasets and/or from smaller scale PD association studies. Meta-analyses on 147 SNPs were supplemented by unpublished GWAS data from up to 16,452 PD cases and 48,810 controls. Eleven loci showed genome-wide significant (P < 5 × 10(-8)) association with disease risk: BST1, CCDC62/HIP1R, DGKQ/GAK, GBA, LRRK2, MAPT, MCCC1/LAMP3, PARK16, SNCA, STK39, and SYT11/RAB25. In addition, we identified novel evidence for genome-wide significant association with a polymorphism in ITGA8 (rs7077361, OR 0.88, P  =  1.3 × 10(-8)). All meta-analysis results are freely available on a dedicated online database (www.pdgene.org), which is cross-linked with a customized track on the UCSC Genome Browser. Our study provides an exhaustive and up-to-date summary of the status of PD genetics research that can be readily scaled to include the results of future large-scale genetics projects, including next-generation sequencing studies.

The contribution of rare coding sequence variants to genetic susceptibility in complex disorders is an important but unresolved question. Most studies thus far have investigated a limited number of genes from regions which contain common disease associated variants. Here we investigate this in inflammatory bowel disease by sequencing the exons and proximal promoters of 531 genes selected from both genome-wide association studies and pathway analysis in pooled DNA panels from 474 cases of Crohn's disease and 480 controls. 80 variants with evidence of association in the sequencing experiment or with potential functional significance were selected for follow up genotyping in 6,507 IBD cases and 3,064 population controls. The top 5 disease associated variants were genotyped in an extension panel of 3,662 IBD cases and 3,639 controls, and tested for association in a combined analysis of 10,147 IBD cases and 7,008 controls. A rare coding variant p.G454C in the BTNL2 gene within the major histocompatibility complex was significantly associated with increased risk for IBD (p = 9.65x10-10, OR = 2.3[95% CI = 1.75-3.04]), but was independent of the known common associated CD and UC variants at this locus. Rare (<1%) and low frequency (1-5%) variants in 3 additional genes showed suggestive association (p<0.005) with either an increased risk (ARIH2 c.338-6C>T) or decreased risk (IL12B p.V298F, and NICN p.H191R) of IBD. These results provide additional insights into the involvement of the inhibition of T cell activation in the development of both sub-phenotypes of inflammatory bowel disease. We suggest that although rare coding variants may make a modest overall contribution to complex disease susceptibility, they can inform our understanding of the molecular pathways that contribute to pathogenesis.