Classical immunogenetics, linkage studies, and genome-wide association studies (GWAS) have identified a set of ~50 chromosomal regions which individually contribute significantly to risk for type 1 diabetes (T1D). Knowledge of the genes in these regions provides an opportunity to probe the molecular underpinnings of the disease, potentially facilitating improved disease prediction and/or novel strategies for prevention or therapeutic intervention. ImmunoChip fine mapping in T1D indicates that the majority of credible causative genetic variants for T1D overlap with tissue specific enhancer (but not promoter) elements, emphasizing the need for a context dependent understanding of the mode of action for genes contributing to T1D pathogenesis through the study of relevant cell types. To address this need, we have, in preliminary studies, processed more than 800 RNA-Seq samples, transcriptionally profiling T and B lymphocyte subsets in both T1D cases and controls. These preliminary studies have yielded three key observations: (1) Genes located in T1D-associated chromosomal regions are significantly enriched for cell-type specific alternative splicing events. These alternative splicing events are observed in 72% of T1D genes but only 32% of genes overall (P<0.0001) and include event types such as intron retention or exon skipping that are likely to have significant effects on the expression and function of the proteins encoded by these genes. (2) Non-coding polymorphisms in regulatory regions that are identified as credible causative variants for T1D, are significantly associated with these alternative splicing events in a highly tissue-specific manner. (3) Genes located in chromosomal regions associated with T1D display correlated expression in lymphocytes, allowing them be clustered into co- expression modules that provide insights into regulation and function. Our preliminary findings suggest that transcript isoforms, and, by implication, the protein isoforms that they encode, are altered in T1D through genetically regulated shifts in exon usage in specific cell types. To gain a comprehensive understanding of the impact of alternative splicing on T1D risk, and to identify and characterize disease relevant transcripts and isoforms, we propose to apply long read sequencing to RNA from lymphocytes in T1D and jointly analyze short and long read data to obtain reliable, cell type specific, information on transcript structure and usage. We will then characterize the effects of alternative splicing and transcript isoform usage on two candidate T1D genes that are part of a co-expression module in lymphocytes, IKZF1 and SIRPG.
Type 1 diabetes arises from the actions, and possible interactions, of multiple genetic and environmental risk factors. Our studies seek to identify and understand the mechanism of action of genetic risk factors for type 1 diabetes in humans. As germline factors, genetic risk variants are present and amenable to study at all times?before, during and after the development of diabetes. Therefore, genetic information can serve as a potential predictive tool as well as provide insights into pathogenesis occurring during the preclinical phase of the disease where preventive therapies might be applied.
|Newman, Jeremy R B; Concannon, Patrick; Tardaguila, Manuel et al. (2018) Event Analysis: Using Transcript Events To Improve Estimates of Abundance in RNA-seq Data. G3 (Bethesda) 8:2923-2940|