Several studies have demonstrated that children with JIA have an increased prevalence of other autoimmune disorders including thyroiditis or celiac disease [63–66]. Furthermore, relatives of children with JIA have an increased prevalence of autoimmunity [7], and this is pronounced in maternal aunts and grandmothers compared to paternal aunts and grandmothers, demonstrating a parent of origin effect [67]. Familial clustering of autoimmunity is also observed in families of adults with RA or idiopathic inflammatory myopathies [68, 69]. Together, these observations suggest that clinically distinct autoimmune phenotypes share common genetic risk factors. Recent genomic studies have confirmed that many genetic variants predispose to multiple autoimmune phenotypes [45, 70, 71]. Emerging data indicate that while some genes confer a predisposition to autoimmunity in general, other disease specific genes and other factors including smoking or infections might influence susceptibility to individual autoimmune diseases. This is illustrated by the observations that mutations in PTPN22 and STAT4 appear to influence susceptibility to multiple autoimmune phenotypes, whereas NOD2 and ATG16L1 variants only seem to influence the risk of IBD. Similarly PADI4 variants are specific to RA, whereas ITGAM variants are specific to SLE [72]. Meta-analyses of genome-wide studies of clinically distinct autoimmune phenotypes will likely enhance our understanding of common pathways that will be targets for treatment across diseases.

In spite of the success in identifying common variants predisposing to rheumatic diseases, a substantial proportion of heritability to rheumatic diseases is still unexplained. Rare disease-associated variants, in part, have been postulated to explain the missing heritability [28]. Current GWAS assays do not effectively capture low frequency and rare variants associated with disease. One common strategy to identify rare variants predisposing to phenotypes of interest is to catalog variants present in affected individuals, since it is anticipated that they will be enriched for such rare variants. Although sequencing costs are rapidly decreasing, sequencing the entire human genome on a large scale is still not cost effective. Hence investigations have focused on sequencing the entire coding exome in individuals and their families to identify disease-causing mutations. Thus, whole-exome sequencing has helped to identify causal variants in extremely early-onset cases of inflammatory bowel disease [73, 74]. In rheumatology, Belot et al. investigated three siblings with juvenile SLE from consanguineous kindred by whole-exome sequencing and identified a functional homozygous mutation in the PRKCD gene which resulted in reduced expression of a protein involved in deletion of autoreactive B cells [75]. These examples show that rare variants with large effects can predispose to complex traits, and it is possible to identify such variants using whole-exome sequencing. In the next decade, whole genome sequencing might complement our efforts to discover additional genomic variation underlying most phenotypes with substantial heritability.