Journal: Cells

Article Title: Shining the Light on Astrocytic Ensembles

doi: 10.3390/cells12091253

Figure Lengend Snippet: Experimental techniques to label and visualize astrocytes.

Article Snippet: In regard to techniques that allow spatial transcriptomics [ , , ], some already commercialized are Spatial Transcriptomics [ ], GeoMx [ ], and CosMx [ ].

Techniques: Labeling, Expressing, Virus, Transgenic Assay, Electroporation, Imaging, Electron Microscopy, Laser-Scanning Microscopy, Confocal Microscopy, Microscopy, Fluorescence, FACS, Affinity Purification, Activity Assay, Activation Assay, Inhibition

Journal: bioRxiv

Article Title: Sub-cellular Imaging of the Entire Protein-Coding Human Transcriptome (18933-plex) on FFPE Tissue Using Spatial Molecular Imaging

doi: 10.1101/2024.11.27.625536

Figure Lengend Snippet: Design of the CosMx Whole Transcriptome Assay . (A) Distribution of sizes of all transcripts targeted by WTx probes is shown as a histogram. All transcripts greater than or equal to 15 kb in length are binned. (B) Gene density plot overlayed on the T2T genome karyotype demonstrates density of probe target regions across the chromosomes. (C) Schematic description of the ISH probe (blue), reporter design and illustration of the RNA assay workflow. The ISH probe consists of the target-binding and readout domains, the former being a 35–50-nt DNA sequence that hybridizes with target RNA, and the latter can be hybridized with a unique reporter. Each Reporter is conjugated with one of four fluorophores, and will be detected as one of four colors (blue, green, yellow or red) in SMI images. Each reporter has a controlled number of 30 dyes with six photocleavable sites to efficiently quench signals by UV illumination and a washing step before each cyclic reporter readout. First, the FFPE slide undergoes standard tissue preparation to expose RNA targets for ISH probe hybridization. Then the sample is assembled into a flow cell and loaded onto an instrument for cyclic readout with 39 sets of reporters. Because each reporter set contains four reporters with four different fluorophores, 156 unique reporters are used in the SMI assay to bind to the different reporter-landing domains on ISH probes. Following each set of reporter hybridization, high-resolution Z-stacked images are acquired followed by cleavage and removal of fluorophores from the reporters before incubation with the next set of reporters. (D) Each target is assigned with a unique 39-digit barcode with 4 “on” spots (labeled as either B,G,Y or R) and 35 “off” spots (labeled as “0”). Each colored letter of the barcode indicates the presence of the reporter that is associated with the target in the specific reporter hybridization round and its color indicates the fluorophore of the hybridized reporter. “0” means that no reporter binds to the ISH probe in that hybridization round, and the target should be silenced or blank in that round of imaging. For each gene, 4 reporters will sequentially bind to the 4 designated reporter landing domains of the ISH probe throughout the 39 rounds of cyclic reporter readout. (E) Representative 260nm chromatograms for reporter assembly and purification are shown. Unpurified 30 dye reporters with Alexa Fluor-647 (top) shown in contrast to purified product (bottom). Each inset box represents what is represented by each chromatogram peak. Each element is hybridized in excess to ensure fully assembled reporters.

Article Snippet: To benchmark the performance of the CosMx WTx assay against the widely used single-cell RNA sequencing (scRNA-seq) platform, we analyzed adjacent tissue sections from the same FFPE colorectal carcinoma block from a 10X certified service provider (Azenta Life Sciences, South San Francisco, CA).

Techniques: Binding Assay, Sequencing, Hybridization, Incubation, Labeling, Imaging, Purification

Journal: bioRxiv

Article Title: Sub-cellular Imaging of the Entire Protein-Coding Human Transcriptome (18933-plex) on FFPE Tissue Using Spatial Molecular Imaging

doi: 10.1101/2024.11.27.625536

Figure Lengend Snippet: WTx assay performance in cell pellet arrays. (A) Three replicate CosMx runs are plotted by total sum of all transcript calls transformed to log10 scale. Replicate 1 and 2 are done at a small imaging scale (37 FOVs), while Replicate 3 is done at a large imaging scale (370 FOVs), suggesting that consistency is not compromised on slides with more FOVs. Blue dots denote transcript calls, orange “X”s denote system controls, and green squares denote negative controls. The black line shows the best fit line to transcript calls (omitting system and negative controls), while the orange line shows a line of slope = 1, and no intercept. (B) CosMx WTx RNA expression profiling is concordant with bulk RNA-seq. Log10 transformed CosMx data, the sum of all transcripts across a single FOV (y-axis), is compared against log10 transformed bulk RNA-seq data, in TPM, collected from the CCLE (x-axis) for 36 cell lines on a 5 µm thick section of FFPE cell pellet array. Red lines describe segmented regression, and orange lines show the log10 transformed estimated breakpoints from each cell line in TPM. (C) Estimated breakpoints for each cell line in TPM, across 12 FOVs (per cell line) and 3 replicate slides. Black lines show 95% confidence interval, and the red line shows median value. (D) Pearson’s correlation between log10 transformed RNA-seq and CosMx data, for all transcripts above the breakpoint, across 12 FOVs (per cell line) and 3 replicate slides. The black line shows 95% confidence interval, and the red line shows median value. (E) Correlation between the log10 transformed transcripts for the full WTx panel (y-axis) and the log10 transformed transcripts for a subset of 6064 targets (x-axis) that corresponds to the commercial 6k panel, with targets selected for further redesign temporarily removed. Black dotted line shows the best fit regression.

Article Snippet: To benchmark the performance of the CosMx WTx assay against the widely used single-cell RNA sequencing (scRNA-seq) platform, we analyzed adjacent tissue sections from the same FFPE colorectal carcinoma block from a 10X certified service provider (Azenta Life Sciences, South San Francisco, CA).

Techniques: Transformation Assay, Imaging, RNA Expression, RNA Sequencing Assay

Journal: bioRxiv

Article Title: Sub-cellular Imaging of the Entire Protein-Coding Human Transcriptome (18933-plex) on FFPE Tissue Using Spatial Molecular Imaging

doi: 10.1101/2024.11.27.625536

Figure Lengend Snippet: Performance comparison between CosMx WTx and single-cell RNA sequencing in FFPE colon cancer tissue. (A) Workflow schematic and throughput for Chromium FLEX and CosMx WTx assays. CosMx enables multi-modal integration of H&E, immunofluorescence and transcriptomics data on the same tissue section. (B) Histograms of detected total transcript counts per cell and gene features per cell from both platforms. (C) Cell type-specific comparison between the platforms (a coarse-level of 7 primary cell types was adopted). (D) Cell compositions of the tissue obtained from Chromium and CosMx were compared. To balance the difference in sample size, the CosMx dataset was subsampled: ∼6,000 cells were randomly chosen and mean composition was calculated (n = 100, mean + s.d.). (E) Cell type composition for entire dataset population from both platforms.

Article Snippet: To benchmark the performance of the CosMx WTx assay against the widely used single-cell RNA sequencing (scRNA-seq) platform, we analyzed adjacent tissue sections from the same FFPE colorectal carcinoma block from a 10X certified service provider (Azenta Life Sciences, South San Francisco, CA).

Techniques: Comparison, RNA Sequencing Assay, Immunofluorescence

Journal: bioRxiv

Article Title: Sub-cellular Imaging of the Entire Protein-Coding Human Transcriptome (18933-plex) on FFPE Tissue Using Spatial Molecular Imaging

doi: 10.1101/2024.11.27.625536

Figure Lengend Snippet: Performance of CosMx WTx in cultured cells. (A) Upfront morphology staining cell membrane (yellow) and nuclei stain (blue) shown in top left. In the image below, cell segmentation based on the above markers shown. Red dashed line in top left image delineates the boundaries of the area shown in the top right image. In the top right image, the red arrow points to one cell of interest (ID 232). Additional red box inset in the top right image shows area for bottom right image. Bottom right image shows 3D localization of transcripts in cells. (B) Assay reproducibility was validated in two independent runs using HEK293T cells, as shown in the upper correlation plot. Correlation with bulk RNA-seq data in the same line is presented in the lower plot. (C) Violin plots showing mean transcript counts and gene features per cell detected in 3 cell lines (n > 50,000 cells per cell line). (D) Upon dimension reduction, data for the three cell lines are presented via UMAP. (E) The top differentially expressed genes across the cell lines are shown with heatmap.

Article Snippet: To benchmark the performance of the CosMx WTx assay against the widely used single-cell RNA sequencing (scRNA-seq) platform, we analyzed adjacent tissue sections from the same FFPE colorectal carcinoma block from a 10X certified service provider (Azenta Life Sciences, South San Francisco, CA).

Techniques: Cell Culture, Staining, Membrane, RNA Sequencing Assay

Journal: bioRxiv

Article Title: Sub-cellular Imaging of the Entire Protein-Coding Human Transcriptome (18933-plex) on FFPE Tissue Using Spatial Molecular Imaging

doi: 10.1101/2024.11.27.625536

Figure Lengend Snippet: Pathway and Ligand-Receptor interactions identify global and spatial intratumor heterogeneity in colon cancer. (A) Major cell types identified using CosMx SMI WTx data for 414,005 cells. Cells are shown in their native XY space and UMAP space. Large tumor nest (black outline) is used for LR evaluation (panel D). (B) Local pathway enrichment scores for select PROGENy pathways. (C) Global view of pathway enrichment for all major cell type classifications. (D) Local LR scores for a given LR complex. Shown here are the eight complexes with the highest region-wide variability.

Article Snippet: To benchmark the performance of the CosMx WTx assay against the widely used single-cell RNA sequencing (scRNA-seq) platform, we analyzed adjacent tissue sections from the same FFPE colorectal carcinoma block from a 10X certified service provider (Azenta Life Sciences, South San Francisco, CA).

Techniques:

Journal: bioRxiv

Article Title: A Comparative Analysis of Imaging-Based Spatial Transcriptomics Platforms

doi: 10.1101/2023.12.13.571385

Figure Lengend Snippet: A. Schematic of the Experimental Design: Serial sections of FFPE prostate adenocarcinoma were simultaneously processed for spatial transcriptomics using Nanostring CosMx SMI and 10x Genomics Xenium, single-nucleus RNA sequencing (snPATHO-seq), and histological analysis. B. Annotated UMAP Embedding of snPATHO-seq Data: Cell types were annotated utilizing reference scRNA-seq data of human prostate from Tuong et al. , along with canonical markers for each cell type. C. Expression and Detection Frequency of Top 3 Markers: The top three markers (p<0.05, ranked by logFC) for each cell type are shown. Log(CP10k+1) represents the log-transformed transcript counts per 10,000 transcripts. D. Top: H&E Stain of Each Section Following Xenium or CosMx Imaging: Post-imaging, CosMx tissues undergo additional processing for immunofluorescence, which affects the stain quality. Bottom: Spatial Transcriptomics Data: Each cell is depicted as a point, positioned at its centroid. Cells are color-coded by total transcript count. E. Correlation of Mean Transcript Counts for Each Probe in the Assay Across Replicates on the Same Platform: The Pearson correlation coefficient for both platforms >0.99.

Article Snippet: Post-run H&E staining of CosMx slides also utilized the precise steps and timing of the 10X Genomics H&E staining protocol, except for the initial sodium hydrosulphite treatment.

Techniques: RNA Sequencing, Expressing, Transformation Assay, Staining, Imaging, Immunofluorescence

Journal: bioRxiv

Article Title: A Comparative Analysis of Imaging-Based Spatial Transcriptomics Platforms

doi: 10.1101/2023.12.13.571385

Figure Lengend Snippet: A. Tissue structure of colon (top; Xenium: public data from 10x Genomics, CosMx: Garrido-Trigo et al. ), lung cancer (middle, Xenium: public data from 10x Genomics, CosMx: public data from NanoString Technologies), and breast cancer (bottom; Xenium and CosMx provided the Suzuki lab, University of Tokyo). Cell annotations are based on automated annotation from scRNA-seq reference datasets. B. Average detection of gene-targeting and control probe sets. C. Spatial autocorrelation (Moran’s I) of all probes in each data set (top panels). A comparison of Moran’s I values for common gene targets is shown in each bottom panel. The dashed line reflects a slope of 1. D. Correlation of spatial transcriptomics data from each platform with a scRNA-seq reference (Breast cancer, Wu et al. ; Lung cancer, Kim et al. ; Colon, custom integration of public data, available with data from this manuscript). Given that reference data is from a different source and cellular composition may differ, values only from annotated endothelial cells are shown as an example. E. Transcript abundance for canonical markers in cells with the specified annotation are colored as the maximal value.

Article Snippet: Post-run H&E staining of CosMx slides also utilized the precise steps and timing of the 10X Genomics H&E staining protocol, except for the initial sodium hydrosulphite treatment.

Techniques: Control, Comparison