Abstract Receptors are essential mediators of cellular physiology, which facilitate molecular and cellular cross-talk with the environment. Nearly 20% of the all known celiac disease (CD) genes are receptors by function. We hypothesized that novel biologically relevant susceptibility receptor genes act in synergy in CD pathogenesis. We attempted to identify novel receptor genes in CD by re-analyzing published Illumina Immunochip dense genotype data for a north Indian and a European (Dutch) cohort. North Indian dataset was screened for 269 known receptor genes. Association statistics for SNPs were considered with minor allele frequency >15% and association P ≤ 0.005 to attend desired study power. Identified markers were tested for cross-ethnic replication in a European CD dataset. Markers were analyzed in-silico to explain their functional significance in CD. Six novel SNPs from MOG (rs29231, p = 1.21e-11), GABBR1 (rs3025643, p = 1.60e-7), OR2H2 (rs1233388, p = 0.0002), ABCF1 (rs9262119, p = 0.0005), ADRA1A (rs10102024, p = 0.003), and ACVR2A (rs7560426, p = 0.004) were identified in north Indians, of which three genes namely, GABBR1 (rs3025643, p = 5.38e-8), OR2H2 (rs1233388, p = 3.29e-5) and ABCF1 (rs9262119, p = 0.0002) were replicated in Dutch. Tissue specific functional annotation, potential epigenetic regulation, co-expression, protein-protein interaction and pathway enrichment analyses indicated differential expression and synergistic function of key genes that could alter cellular homeostasis, ubiquitination mediated phagosome pathway and cellular protein processing to contribute for CD. At present multiple therapeutic compounds/drugs are available targeting GABBR1 and ADRA1A, which could be tested for their effectiveness against CD in controlled drug trials. Subject terms: Genome-wide association studies, Coeliac disease Introduction Celiac disease (CD) is a chronic autoimmune disease triggered by gluten protein in genetically susceptible individuals. The diagnosis of CD shows an increase in intraepithelial lymphocytes and altered crypt:villus ratio^[34]1. Although, prevalence of CD varies from 0.5% to 1% in different parts of the world, etiology of this complex disease is not completely known. HLA DQ2/8 haplotypes are the major predisposing genetic factors of CD, which is corroborated by twin studies and several GWAS studies^[35]2,[36]3. GWAS on CD identified a total of 74 SNPs from 56 non-HLA loci, of which estimated 28 loci potentially alter the immune-related direct response that underlies the CD pathogenesis while several others regulate various associated biological processes and pathways. Notably, known 16 SNPs (~20%) associated with CD localized within or near to the genes with receptor function. This potentially highlighted the substantial contribution of receptor genes in CD. The dietary gluten upon digestion is broken down into smaller proteins (glutenin and gliadin), which are absorbed in the small intestine and might trigger some abrupt cell signaling that initiate CD pathogenesis cascade. Cell-cell communication occurs via cascade of mechanisms including activation of receptors, transducers and enzymes. In gastrointestinal (GI) tissue smooth muscle contraction initiates with Ca^2+ influx via L-type Ca^2+ channels, which are activated through G-protein coupled receptors (GPCRs)^[37]4,[38]5. Ca^2+ sensitivity leads to inflammation induced hypocontractility and increased expression of cytokines. Cytokines binding to their receptors affects intracellular signaling and expression of GPCR, which can influence Ca^2+ sensitization. However, in case of clinical inflammatory bowel disease (IBD) and experimental models of colitis, cytokines generated by Th17 cells are prominent in microbial infections, ischemia/reperfusion injury^[39]5. Notably, IL15, an unique cytokine is up regulated by stress and inflammation on any cell type but its expression requires binding with IL15R^[40]6. Another receptor NOTCH4 has been identified to stimulate NFkB signaling and inhibits E-cadherin, which is critical for gut epithelial biology^[41]7. Existing molecular evidences and recent genetic studies highlighted the significant contribution of receptor genes in CD pathogenesis. Further genetic studies are warranted to identify and/or replicate receptor genes, which may help in identifying the environmental triggers and their cross talk with the cellular components for the onset of CD. In the present study we performed a hypothesis driven association study to identify novel receptor gene(s). Immunochip genotype data were available in-house for a total of 1227 north Indian subjects including 497 CD patients and 736 healthy controls^[42]8,[43]9. Replication of identified risk variants was done using a European CD association summary statistics data, available to us. This study identified six novel risk variants from receptor genes, which are functionally relevant in CD pathogenesis. Based on the findings of this study a molecular model has been proposed which demonstrate the synergistic contribution of the identified novel genes. Several known therapeutic molecules were identified, which may be screened and tested in drug trials for their effectiveness in treating CD. Methods Strategy to select gene of interest A manually curated exhaustive list of receptor genes was made. Receptors were defined as known cellular protein that can bind to a specific ligand and initiate specific signaling cascade. KEGG and Gene Ontology (GO) pathway databases were used to curate a list of known receptor genes for his study^[44]10,[45]11. Irrespective of their site of expression, pathways, cellular localization, all the genes with receptor functions were screened. Gene functions were confirmed by NCBI-Gene and RCSB-PDB (PDB). Alternative gene names and putative gene ids were also cross-checked in NCBI-Gene and Ensembl to avoid the possibility of any gene of interest to be missed/go unnoticed. Further, to reconfirm and validate their molecular and cellular function Uniprot and Gene Cards databases were used^[46]12,[47]13. Genotype data and association statistics Previously published association summary statistics of north Indian celiac disease study (using Illumina Immunochip platform) was available in-house. This dataset was used to identify novel receptor genes associated with CD. Majority of CD association studies were conducted among groups with European ancestry, therefore, we replicated the identified CD associated risk variants in a published European association study dataset. A Dutch CD association study genotypes were obtained from Prof. Cisca Wijmenga’s laboratory, University Medical Center, Groningen, The Netherlands. Details of the population and genotype data characteristics are available elsewhere^[48]8,[49]14. Illumina Immunochip annotation file (hg19) was used to systematically pull out genetic variations (SNPs) mapped against all the receptor genes enlisted in the previously described exhaustive list of receptor genes. These genetic markers include exonic, intronic, UTRs and intergenic variations falling within the index genes as given in the Immunochip annotation file. Genomic coordinates of each of these markers were converted as per hg38 using UCSC LiftOver browser. Marker ids were upgraded as given in the 1000 Genomes database. Summary statistics comprising of reference/minor alleles, minor allele frequency (MAF), association p-value and odds ratio (OR) with 95% CI were documented for each of the SNPs. For further analyses we selected all the SNPs with association p-value < 0.005, having minor allele frequency >15% to obtain approximately 90% study power to detect true associations. Level of significance was kept modest to identify and investigate the cumulative functions of biologically significant receptor genes and reduce chance to miss out any. Association signals identified among north Indian population were tested for cross-ethnic replication in a European celiac disease association study dataset mentioned above^[50]8,[51]14. Associations were evaluated for SNPs originally identified in north Indians population. Conventional 5% level of significance was used to determine the replication of each of these SNPs. To know more about the importance of identified genes in human diseases biology, cross-disease association was checked by scanning all the GWAS reports till date. Relevant data were extracted from NHGRI-EBI GWAS catalog ([52]https://www.ebi.ac.uk/gwas/), which is an up-to-date public database to access and scan all the reported GWAS. In-silico functional annotation and gene prioritization Genetic annotation of SNPs was done using RefSeq references. Molecular