The impact of proinflammatory cytokines on the β-cell regulatory landscape provides insights into the genetics of type 1 diabetes

This collection provides multi-omics datasets generated to investigate the effects of pro-inflammatory cytokines (IFNγ and IL-1β) on pancreatic β-cell regulation and function. The resource includes processed and statistically analyzed results from RNA-seq (human islets and EndoC-βH1 cells), UMI-4C, and DNA methylation (EPIC array) experiments. Together, these datasets capture cytokine-induced changes in gene expression, chromatin interactions, and DNA methylation, offering an integrated view of β-cell responses to inflammatory stress. For more information refer to the original publication: Ramos-Rodríguez, M., Raurell-Vila, H., Colli, M.L. et al. The impact of proinflammatory cytokines on the β-cell regulatory landscape provides insights into the genetics of type 1 diabetes. Nat Genet 51, 1588–1595 (2019). https://doi.org/10.1038/s41588-019-0524-6 RNA-seq data This dataset contains differential expression results from human pancreatic islets and EndoC-bH1 cells exposed to pro-inflammatory cytokines (IFNγ + IL-1β) for 48 h, compared to untreated controls. Two objects are provided: BLK_CYT_RNA_HI: differential expression results for human islets. BLK_CYT_RNA_EC: differential expression results for EndoC-βH1 cells. Read counts generated with htseq-count were used as input for differential analysis with DESeq2 v1.24.0, applying a paired-sample design. Genes were classified as differentially expressed when showing an adjusted p-value (FDR) < 0.05 and an absolute log2 fold change > 1; genes not meeting these criteria were labeled as stable. Each file is an RData object containing a GRanges object with genomic coordinates and metadata columns, including gene identifiers, external gene names, biotypes, base mean expression, log2 fold change, standard error, test statistics, p-values, adjusted p-values, and classification type (stable or differential). UMI-4C Data has been reprocessed using the UMI4Cats R package (doi: 10.1093/bioinformatics/btab392). The pipeline includes read demultiplexing, alignment and UMI-based deduplication, followed by the generation of contact profiles and smoothing across restriction fragments. Interaction windows were defined with makeWindowFragments(). Windows overlapping regulatory elements were used as input to identify differential contacts between cytokine-treated and control samples using Fisher’s exact test (fisherUMI4C()), filtering out windows with fewer than 20 UMIs. Infinium MethylationEPIC array DNA was extracted from EndoC-bH1 cultures that were either exposed or unexposed to the pro-inflammatory cytokines IL-1β and IFN-γ for 48 hours (n = 5 biological replicates per condition). Raw array data were preprocessed and analyzed using the RnBeads R package (v3.2.0). Differential methylation between cytokine-treated and control samples was assessed using empirical Bayesian hierarchical linear models implemented in the limma package (v3.40.0). Analyses were performed on M values (log-transformed β values). P-values were adjusted for multiple testing using the Benjamini–Hochberg false discovery rate (FDR) correction. CpG sites were considered differentially methylated if they exhibited an FDR-adjusted p-value < 0.05 and an absolute β-value difference > 0.2 (≥ 20% methylation change). A full list of differentially methylated CpGs is provided in Supplementary Table 7 of the associated publication.