ABSTRACT Sjögren's Syndrome (SS) is a systemic autoimmune disease affecting the exocrine system, with hallmark symptoms of dry mouth and/or dry eyes caused by the dysfunction of salivary and/or lacrimal glands, respectively. While Genome-Wide Association Studies (GWAS) and other studies have increased our knowledge of genetic risk factors for SS, the disease etiology remains not well understood, and such risk factors have not been translatable to any immunological treatment options for SS. The NIH-NIDCR-funded Sjögren's International Collaborative Clinical Alliance (SICCA) was established to improve the understanding, diagnosis and treatment of patients with SS by developing/validating standardized classification criteria for SS; and developing a rich biospecimen repository with clinical data to be used for future epidemiologic, pathogenesis, and genetic studies of SS.[1, 2] For this project, we will focus on genomic data and measures of the 2016 ACR-EULAR classification criteria, involving ocular, oral, and autoantibody manifestations. As shown in our previous work, the genetics of SS varies with ancestry; thus, we will cluster patients by both the criteria subphenotypes and genetic ancestry. We believe that accounting for disease heterogeneity in this way will enable us to more precisely identify disease pathways and mechanisms. Using previously secured funding, we are completing DNA methylation typing on LSGs in 373 SICCA patients and single-cell RNA sequencing (scRNAseq) on PBMCs of 86 SICCA patients who also have DNA methylation profiling. This data provides a unique opportunity for multi-omics analysis to determine correlates between LSG tissue epigenetics, peripheral blood cell-type distributions and cell-specific gene expression by SS subsets. First, using GWAS data and DNA methylation data from LSG biopsies, we will identify genetic and epigenetic modifications associated with subtypes of SS in SICCA patients. We will then examine the relationships between them by testing for genotype-specific methylation and expression, and utilizing mendelian randomization and causal inference testing to investigate causality between these measures. Second, we will analyze scRNAseq data to identify how cell types, states and cell- specific gene expression correlate with SS subtypes. Finally, we will integrate genetics, epigenetics, and transcriptomics to determine multi-omics profiles associated with SS subtypes. We will jointly model associated features from the genomic data to investigate causal pathways via correlation networks, conditional analysis, and machine learning. We anticipate that SS subtypes will exhibit specific relationships within the multi- omics data and that this will advance our understanding of SS disease processes, leading to better treatment targets.