Journal: Molecular and Cellular Endocrinology

Article Title: Stem/progenitor cells in normal physiology and disease of the pancreas

doi: 10.1016/j.mce.2021.111459

Figure Lengend Snippet: Transcriptomic profiling of human and embryonic tissues, experimental animal models and patient-derived cell lines via scRNA-seq enables the study of pancreas progenitors. (1) Single-cell suspensions enter high throughput microfluidic systems (e.g. 10X Chromium, Drop-Seq) allowing thousands of cells to be processed and sequenced. (2) Cells undergo dimensionality reduction and are clustered based upon expression profiles – represented via UMAP or t-SNE - enabling the identification of novel cell types and investigation of cellular heterogeneity. (3) Following clustering, differential expression analysis reveals changes in gene expression across cell types. (4) Prediction of cell trajectories can be inferred based upon changes in gene expression over a ‘pseudo’ time-course. Cells are ordered in a 2D space based upon the closeness of their expression pro_les. Overlay of a minimal spanning tree (MST) identifies the longest continual path linking these cells – uncovering cell lineages. (5) Individual trajectories can be dissected and changes in specific gene expression changes plotted in both a supervised and unsupervised manner (6) The development of algorithms (e.g. StemID, SCENT) has enabled the prediction of cell clusters with high potency, stem-like features. Used in conjunction with pseudotime analysis, these algorithms can infer a starting point of differentiation trajectories, as well as identifying novel stem cells in adult tissues.

Article Snippet: Transcriptomic profiling of human and embryonic tissues, experimental animal models and patient-derived cell lines via scRNA-seq enables the study of pancreas progenitors. (1) Single-cell suspensions enter high throughput microfluidic systems (e.g. 10X Chromium, Drop-Seq) allowing thousands of cells to be processed and sequenced. (2) Cells undergo dimensionality reduction and are clustered based upon expression profiles – represented via UMAP or t-SNE - enabling the identification of novel cell types and investigation of cellular heterogeneity. (3) Following clustering, differential expression analysis reveals changes in gene expression across cell types. (4) Prediction of cell trajectories can be inferred based upon changes in gene expression over a ‘pseudo’ time-course.

Techniques: Derivative Assay, High Throughput Screening Assay, Expressing

Journal: bioRxiv

Article Title: Single-Cell Data Integration and Cell Type Annotation through Contrastive Adversarial Open-set Domain Adaptation

doi: 10.1101/2024.10.04.616599

Figure Lengend Snippet: A) Comparison of SAFAARI’s performance with the selected reference-based cell-type annotation models in both open-set and closed-set settings. The scRNA-seq data from eight different tissues in the Tabula Muris cell atlas was obtained where the gene counts were derived using two techniques: 10x Genomics and FACS-based cell capture in plates (FACS). For the performance assessment, either FACS or 10x was considered as the source dataset, and the other as the target dataset, to evaluate reference-based cell type annotation or label transfer in the presence of a technology-based domain-shift or batch effect. Two scenarios were considered: the closed-set, where only cell types common to both source and target datasets were included, and the open-set, where the target dataset contained an unknown cell type not present in the source dataset . B) Heatmap representing the confusion matrix across eight tissues (target: FACS), showing cell-type-specific annotation performance. Columns represent the actual cell labels, while rows show the predicted cell labels. The cell type coloured in navy blue represents the unknown cell type whose instances were removed from the source dataset. Colours in the viridis palette and indicate the proportion of cells relative to the sum of the column (i.e., values across columns should add up to 1.0). This represents the proportion of correct classifications (diagonal values) and misclassifications for each particular cell type represented by the column names. C) UMAP of open-set Label transfer result of SAFAARI on four human pancreas datasets generated with different technologies, including microfluidic (Fluidigm C), droplet-based (InDrops) and plate-based scRNA-seq (CEL-seq2, Smart-seq2) as detailed in . It demonstrates SAFAARI’s superior batch mixing, cell separation and unknown cell type detection.

Article Snippet: These methods range from microfluidic droplet-based platforms (such as 10x Genomics Chromium, Drop-seq, and inDrops) to plate-based scRNA-seq technologies like Smart-seq, Smart-seq2, and Smart-seq3, resulting in substantial heterogeneity across datasets.

Techniques: Comparison, Derivative Assay, Generated

Journal: Frontiers in Oncology

Article Title: Unlocking ovarian cancer heterogeneity: advancing immunotherapy through single-cell transcriptomics

doi: 10.3389/fonc.2024.1388663

Figure Lengend Snippet: Summary of recent findings of scRNA-seq in Ovarian cancer.

Article Snippet: , Metastatic HGSOC/malignant tumor biopsy , N=6 9,885 single cells from omental tumor samples of 6 patients , Pre-chemotherapy = 2/6 Post-chemotherapy = 4/6 , 1. Identification of unique sub-populations of CD274 + and IRF8 + macrophages, CD4 + GNLY + T cells, plasmablasts and plasma B cells. 2. Transcriptional analysis of immune cells stratifies our patient samples into 2 groups: (1) high T cell infiltration and (2) low T cell infiltration. 3. Plasmablast and plasma B cell clusters, and NR1H2 + IRF8 + and CD274 + macrophage clusters, suggesting an anti-tumor response in the high Tinf group. , Drop-seq microfluidic, Illumina’s NextSeq 500 , ( ) .

Techniques: Phospho-proteomics, Activity Assay, Gene Expression, Marker, Multiplex Assay, Clinical Proteomics, RNA Sequencing, Sequencing, Activation Assay, Expressing, Migration, Functional Assay