biostudies-arrayexpressJohannes BalkenholHomo sapiensScanpy (v1.10), scFates, Palantir10x Genomics Chromium Software Suite (v6.1)Cell Ranger (v6.1), STAR (v2.7.9a)-https://www.ebi.ac.uk/biostudies/studies/E-MTAB-17360Study Description for ArrayExpress Submission: This study investigates cellular heterogeneity and mechanisms of gemcitabine resistance in pancreatic cancer using single-cell RNA sequencing of PANC-1 cells cultured in a 3D environment (and 2D control). Cells were seeded onto a decellularized porcine intestinal matrix to mimic the tumor microenvironment. The experiment analyzed transcriptional profiles before gemcitabine treatment, with particular attention to invasion and resistance states induced by TGFβ. Single-cell RNA-seq was used to map the transcriptional landscape of 3D pancreatic tumor tissues, capturing cell state transitions, differentiation trajectories, and resistance-associated profiles. The analysis identified cell cycle–dependent resistance signatures (late S to G2(/M) phase), tissue differentiation-associated states, and TGFβ-induced epithelial-to-mesenchymal transition (EMT) pathways contributing to chemoresistance. The dataset includes untreated samples with and without TGFβ stimulation, enabling identification of attractor states and transitional paths leading to gemcitabine resistance. This study provides insight into tumor cell plasticity and dedifferentiation in a 3D matrix environment, supporting early diagnostic and therapeutic research in pancreatic cancer.biostudies-arrayexpressSequencing - Title: Illumina NovaSeq 6000 paired-end sequencing of 10x Genomics libraries Description: Libraries were sequenced on an Illumina NovaSeq 6000 instrument across two lanes using a paired-end dual-index run: 28 bp Read 1 (16 bp cell barcode + 12 bp UMI), 90 bp Read 2 (transcript sequence), and 10 bp i7 and 10 bp i5 sample indices. Both the gene-expression (GEX) and the CellPlex multiplexing-capture (CP) libraries were sequenced in the same run. BCL files were converted to FASTQ using Cell Ranger mkfastq. Data quality was verified using FastQC and Cell Ranger summary metrics.Nucleic Acid Extraction - Title: Single-cell preparation, CellPlex labelling and GEM generation for 10x Genomics Chromium Description: Single-cell suspensions were prepared by enzymatic dissociation and filtration through 40 µm strainers; cell viability exceeded 85 %. Each of the eight biological samples was labelled individually with a 10x Genomics CellPlex cell multiplexing oligo (CMO), washed to remove unbound CMO, counted, and combined in equal proportions into a single pooled suspension. Approximately 10,000 viable cells from the pool were loaded into a 10x Genomics Chromium Controller (Single Cell 3′ v3 chemistry). Cells were lysed within Gel Bead-in-Emulsion (GEM) droplets, and both mRNA and CMO molecules were captured by barcoded oligo-dT primers attached to the gel beads. No bulk RNA extraction was performed prior to library construction.Sample Treatment - Title: Gemcitabine and TGF-β1 treatments in 3D PANC-1 cultures Description: 3D PANC-1 cultures were treated with gemcitabine (10 µM, 24 h), recombinant human TGF-β1 (10 ng/mL, 48 h), or the combination thereof. Control scaffolds were treated with vehicle (DMSO) under identical conditions. After treatment, cells were processed immediately for single-cell capture.Growth Protocol - Title: Maintenance and expansion of PANC-1 cells Description: PANC-1 cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) containing 10 % fetal bovine serum (FBS) and 1 % penicillin–streptomycin under standard incubator conditions (37 °C, 5 % CO₂). Cells were subcultured at 70–80 % confluence.Sample Collection - Title: 3D culture of PANC-1 pancreatic cancer cells on SISmuc scaffolds Description: PANC-1 human pancreatic ductal adenocarcinoma cells were cultured in 3D on decellularized porcine small intestinal submucosa (SISmuc) scaffolds as described in Dandekar et al. (2023). Scaffolds were seeded with PANC-1 cells and maintained under standard conditions (37 °C, 5 % CO₂, DMEM supplemented with 10 % FBS and 1 % penicillin–streptomycin). At experimental endpoints, cells were recovered from 3D scaffolds using collagenase digestion and washed in PBS prior to single-cell suspension preparation. Parameters: Organism: Homo sapiens Cell line: PANC-1 Culture type: 3D SISmuc (decellularized porcine jejunum) Medium: DMEM + 10 % FBS Temperature: 37 °C CO₂: 5 %Library Construction - Title: 10x Genomics Chromium Single Cell 3′ v3 library construction with CellPlex multiplexing Description: Libraries were generated using the 10x Genomics Chromium Single Cell 3′ Reagent Kit v3 in combination with the 3′ CellPlex Kit, according to the manufacturer's instructions. Eight biological samples were labelled with cell multiplexing oligos (CMO301, CMO302, CMO304–CMO309; CMO303 was not used) and pooled prior to GEM generation. Reverse transcription within GEMs tagged each cDNA and CMO molecule with a unique cell barcode and UMI. Following cDNA amplification, the amplified material was split to construct two separate libraries: a 3′ gene-expression (GEX) library, built by fragmentation, end repair, A-tailing, adapter ligation and sample-index PCR; and a multiplexing-capture (CP) library, generated from the CMO-derived fraction by sample-index PCR. Both libraries were quantified by Bioanalyzer and qPCR prior to sequencing.OrganizationMINSEQE ScoreAssays and DataProcessed DataMAGE-TAB FilesData Transformation - Title: scRNA-seq data quality control, normalisation and downstream processing Description: Raw sequencing reads were processed with Cell Ranger (10x Genomics, v6.1) against the GRCh38 reference (refdata-gex-GRCh38-2020-A). The resulting gene–cell count matrices were imported into Scanpy (v1.10) for quality control, including filtering of low-quality cells and doublet detection with Scrublet. Counts were normalised and log-transformed, highly variable genes selected, and dimensionality reduction, Leiden clustering, and UMAP/Force-Atlas visualisation performed. Pseudotime and trajectory inference were carried out using scFates and Palantir.Sequence Alignment - Title: Alignment and demultiplexing with Cell Ranger multi Description: FASTQ files were processed with Cell Ranger multi (v6.1) against the GRCh38 2020-A reference transcriptome. Cell Ranger multi jointly processed the gene-expression (GEX) and CellPlex multiplexing-capture (CP) libraries, demultiplexing the CMO tags to assign each cell barcode to its biological sample, and generated per-sample gene–cell count matrices. Reads were aligned with STAR (v2.7.9a) within Cell Ranger and filtered for valid UMIs and cell barcodes. QC metrics were reviewed using the Cell Ranger web summary.UnknownTranscriptomicsGenomicsProteomicsCell culture incubator (37 °C, 5 % CO₂)Biological safety cabinet, centrifuge, 40 µm strainerLinux server (GPU/CPU cluster)Illumina NovaSeq 6000Standard cell culture setupRNA-seq of coding RNA from single cellsHomo sapiensJohannes BalkenholfalseSystematic Single-Cell Dissection of Cell Cycle and TGFβ-Induced State Transitions Underlying Gemcitabine Resistance in 3D Pancreatic Tumor TissueStudy Description for ArrayExpress Submission: This study investigates cellular heterogeneity and mechanisms of gemcitabine resistance in pancreatic cancer using single-cell RNA sequencing of PANC-1 cells cultured in a 3D environment (and 2D control). Cells were seeded onto a decellularized porcine intestinal matrix to mimic the tumor microenvironment. The experiment analyzed transcriptional profiles before gemcitabine treatment, with particular attention to invasion and resistance states induced by TGFβ. Single-cell RNA-seq was used to map the transcriptional landscape of 3D pancreatic tumor tissues, capturing cell state transitions, differentiation trajectories, and resistance-associated profiles. The analysis identified cell cycle–dependent resistance signatures (late S to G2(/M) phase), tissue differentiation-associated states, and TGFβ-induced epithelial-to-mesenchymal transition (EMT) pathways contributing to chemoresistance. The dataset includes untreated samples with and without TGFβ stimulation, enabling identification of attractor states and transitional paths leading to gemcitabine resistance. This study provides insight into tumor cell plasticity and dedifferentiation in a 3D matrix environment, supporting early diagnostic and therapeutic research in pancreatic cancer.2026-07-10T00:00:00Z2026-07-23T13:35:06.274Z2026-07-23T13:34:49.781ZE-MTAB-17360ERP202599EFO_0002944EFO_0004170EFO_0003789EFO_0005684EFO_0004917EFO_0005518EFO_0003816EFO_0004184EFO_0003969