biostudies-arrayexpressTobi AlegbeHomo sapiensCellRanger, CellBender, Scrublethttps://www.ebi.ac.uk/biostudies/studies/E-MTAB-16999This scRNA-seq data resource contains various clustered and auto-annotated anndata, as well as associated individual and sample metadata, from IBDverse. It consists of over 1.1 million cells from terminal ileal biopsies of 111 patients with Crohn’s disease and 232 healthy controls.biostudies-arrayexpressSequencing - Libraries were sequenced using a HiSeq4000 sequencer (Illumina; NCD=20, Ncontrol=4) or NovaSeq S4 XP sequencer (Illumina; NCD=91, Ncontrol=228) with 100bp paired-end reads, targeting 50,000 reads per cell.Nucleic Acid Extraction - Terminal ileal biopsies were dissociated on ice to release all major intestinal cell types present in the biopsy (epithelial, immune, and stromal) without stressing the cells and to minimize de novo transcription. 2-3 pinch biopsies were transferred to Hank's Balanced Salt Solution HBSS-/- containing 2 mM EDTA, 0.26 U/µl serine endoprotease isolated from Bacillus licheniformus (Sigma, P5380), 5 µM QVD-OPh (Abcam, ab141421), and 50 µM Y-27632 dihydrochloride (Abcam, ab120129). The biopsies were mechanically minced and gently pipetted every ten minutes during a 30 minute incubation period on ice . Cells were washed with 1% BSA HBSS-/-,centrifuged at 350 x g at 4°C for 5 minutes, and then incubated for 10 minutes at room temperature in HBSS with Mg2+, Ca2+, and without phenol red—including 5 mM CaCl2, 1.5U/µl collagenase IV (Worthington, LS004188), and 0.1 mg/ml DNase I (Stem Cell Technologies, 07900). Next, 0.5M EDTA was added, cells were filtered (30 µm; CellTrics 04-0042-2316), washed, and centrifuged before being incubated for 3 minutes at room temperature in red blood cell lysis buffer (ACK lysis buffer; Gibco, A10492). Cells were washed and centrifuged twice before a final filtration (40 µm; Falcon, 352340) and manual cell counting (haemocytometer, NanoEnTek, DHC-N01). For some samples, we isolated the cells released in the buffer post mechanical mincing (fraction 1) and mixed them at different ratios with cells isolated from the remaining tissue chunks post enzymatic digestion (fraction 2), in an attempt to make the representation of immune cell types more equal to epithelial cells in variably inflamed samples. Fraction 1 and 2 were processed the same way throughout the protocol, except that fraction 1 did not undergo the first serine endoprotease enzymatic digestion step.Library Construction - Single-cell RNA sequencing was undertaken using 3’ 10X Genomics kits (v3.0 and v3.1) according to the manufacturer’s instructions. All samples sequenced under kit version v3.1 had dual indexes, samples sequenced under kit version v3.0 had either single or dual indexes.Sample Collection - Individuals undergoing routine endoscopic assessment were recruited at Addenbrooke’s hospital, Cambridge, UK. Clinical information and metadata for the participants are provided in Zenodo and EGA. All CD participants classified as ‘inflamed’ had a confirmed history of CD and macroscopic evidence of terminal ileal inflammation from tissue sampled during the biopsy. All control participants were undergoing endoscopic assessment or surveillance for healthy and non-cancer related reasons (e.g., history of iron deficiency anaemia, family history of colorectal cancer). Control participants did not have macroscopic evidence of intestinal inflammation, a personal history of cancer, and were not in receipt of corticosteroids or any other immune modulating therapy. Pinch-biopsies of the terminal ileum were collected from all participants and deposited into pre-chilled Hanks Balanced Salt Solution (HBSS) without Mg2+, Ca2+, or phenol red. Samples were placed on ice and immediately transferred to the Sanger Institute.OrganizationMINSEQE ScoreAssays and DataProcessed DataMAGE-TAB FilesData Transformation - CellRanger v7.2.0 was used to demultiplex reads, align reads to GRCh38 with Ensembl version 93 transcript definitions (GRCh38-3.0.0 reference file distributed by 10X Genomics), and generate cell by gene count matrices. CellBender v2.1 55 was then applied to identify droplets containing cells and adjust the raw counts matrix for background ambient transcript contamination. Multiplets were identified and removed using scrublet v0.2.1 57, simulating 100,000 multiplets and calculating the multiplet threshold using the threshold_li function from the scikit-image package v0.17.2 58, initialised using the threshold_otsu function.MetabolomicsUnknownTranscriptomicsGenomicsProteomicsIllumina HiSeq 400010X Genomics ChromiumRNA-seq of coding RNA from single cellsHomo sapiensMonika KrzakCarl AndersonTobi AlegbeTim RainefalseIBDverse: Single-Cell RNA Sequencing of Terminal Ileal Biopsies Identifies Signatures of Crohn’s Disease PathogenesisThis scRNA-seq data resource contains various clustered and auto-annotated anndata, as well as associated individual and sample metadata, from IBDverse. It consists of over 1.1 million cells from terminal ileal biopsies of 111 patients with Crohn’s disease and 232 healthy controls.2026-04-28T00:00:00Z2026-04-30T16:33:38.065Z2026-04-28T13:11:01.833ZE-MTAB-16999S-BSST2944E-MTAB-16998EFO_0002944EFO_0004170EFO_0005684EFO_0005518EFO_0003816EFO_0004184