Title Revealing the Key Regulators of Cell Identity in the Human Adult Pancreas [scRNA-seq] Sample organism Homo sapiens Experiment type Expression profiling by high throughput sequencing
Third-party reanalysis
Summary Cellular identity during development is under the control of transcription factors that form gene regulatory networks. However, the transcription factors and gene regulatory networks underlying cellular identity in the human adult pancreas remain largely unexplored. Here, we integrate multiple single-cell RNA-sequencing datasets of the human adult pancreas, totaling 7393 cells, and comprehensively reconstruct gene regulatory networks. We show that a network of 142 transcription factors forms distinct regulatory modules that characterize pancreatic cell types. We present evidence that our approach identifies key regulators of cell identity in the human adult pancreas. We predict that HEYL, BHLHE41 and JUND are active in acinar, beta and alpha cells, respectively, and show that these proteins are present in the human adult pancreas as well as in human induced pluripotent stem cell (hiPSC)-derived islet cells. Using single cell transcriptomics, we found that JUND represses beta cell genes in hiPSC-alpha cells. Both BHLHE41 and JUND depletion seemed to increase the number of sc-enterochromaffin cells in hiPSC-derived islets. The comprehensive gene regulatory network atlas can be explored interactively online. We anticipate our analysis to be the starting point for a more sophisticated dissection of how transcription factors regulate cell identity in the human adult pancreas. Furthermore, given that transcription factors are major regulators of embryo development and are often perturbed in diseases, a comprehensive understanding of how transcription factors work will be relevant in development and disease.
Overall design Raw reads for five publicly available scRNA-seq datasets (GEO: GSE86469, GEO: GSE81608, GEO: GSE83139, GEO: GSE81547, ArrayExpress: E-MTAB-5061) were downloaded from SRA using SRA toolkit (v2.9.4). Afterwards, reads were aligned to the human reference genome GRCh38.95 using STAR (v2.5.3a) with default parameters followed by the conversion to the coordinate sorted BAM format. Next, the featureCounts command from the “Rsubread” (v1.5.2) package in R (v3.6.1) was used to assign mapped reads to genomic features. Low quality transcriptomes with a mitochondrial contamination greater than 5% and less than 200 expressed genes per cell were excluded from subsequent analyses. The resulting raw count matrix was batch corrected using the FindIntegrationAnchors and IntegrateData functions from the “Seurat” package (v3.1.1). The integrated single-cell RNA-seq data and pySCENIC results were used to create a loom file which can be uploaded to SCope. The embedding of the regulon and integrated gene expression based UMAP clustering were added to the loom file.
Gene regulatory networks were inferred using pySCENIC (python implementation of SCENIC, v0.9.15) in Python version 3.6.9. Integrated read counts were used as input to run GENIE3 which is part of arboreto (v0.1.5). GRNs were subsequently inferred using pySCENIC with the hg38_refseq-r80 motif database and default settings. To control for the stochasticity, which is inherent to pySCENIC, a consensus GRN was generated by merging results from five repeat pySCENIC runs. If regulons were identified in multiple pySCENIC runs, only the regulon with the highest AUC value was retained.
Contributor(s) Vanheer L , Fantuzzi F , To SK , Schiavo AA , Van Haele M , Haesen T , Yi X , Janiszewski A , Chappell J , Rihoux A , Sawatani T , Roskams T , Pattou F , Kerr-Conte J , Cnop M , Pasque V Citation missing Has this study been published? Please login to update or notify GEO . Submission date Aug 19, 2020 Last update date Nov 22, 2022 Contact name Vincent Pasque E-mail(s) vincent.pasque@kuleuven.be
Organization name KU Leuven
Department Development and Regeneration
Street address Herestraat 49 bus 804
Leuven ZIP/Postal code Country Belgium This SubSeries is part of SuperSeries:

GSE218548

Revealing the Key Regulators of Cell Identity in the Human Adult Pancreas

Download family	Format
SOFT formatted family file(s)
MINiML formatted family file(s)	MINiML
Series Matrix File(s)

Supplementary file	Download	File type/resource
GSE156490_Merged_raw_countData.tsv.gz	357.1 Mb	(ftp) (http)
GSE156490_countData_all_t.tsv.gz	1.7 Gb	(ftp) (http)
GSE156490_scPancreasAtlas.loom.gz	788.2 Mb	(ftp) (http)
Raw data are available in SRA
Processed data are available on Series record