Apr 29, 2025
Loupe Browser Walkthrough Visium HD Spatial Gene Expression Library, Human Colorectal Cancer (FFPE)
- Jennifer Becker1
- 1University of Rochester Medical Center
Protocol Citation: Jennifer Becker 2025. Loupe Browser Walkthrough Visium HD Spatial Gene Expression Library, Human Colorectal Cancer (FFPE). protocols.io https://dx.doi.org/10.17504/protocols.io.rm7vz1e25lx1/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: April 28, 2025
Last Modified: April 29, 2025
Protocol Integer ID: 189850
Abstract
This protocol is intended to serve as an example of the capabilities of 10X Genomics Loupe Browser within their hosted Visium HD FFPE Colorectal Cancer dataset. More information on this dataset can be found at https://www.10xgenomics.com/datasets/visium-hd-cytassist-gene-expression-libraries-of-human-crc.
Introduction
Introduction
This protocol is intended to serve as an example of the capabilities of 10X Genomics Loupe Browser within their hosted Visium HD FFPE Colorectal Cancer dataset. More information on this dataset can be found at https://www.10xgenomics.com/datasets/visium-hd-cytassist-gene-expression-libraries-of-human-crc.
Importing Data
Importing Data
Importing data into Loupe browser is as easy as navigating to your cloupe file!
Visualize Capture Area and Clustering
Visualize Capture Area and Clustering
Once you've loaded the cloupe file the microscope image (in this case, a brightfield H&E image) and HD gene expression data are visualized together in the center of the screen.
You can easily zoom in and out by clicking on the magnifying glass or simply scrolling with your mouse. The scale bar in the lower right hand corner is adjusted accordingly.
We can also adjust the projection settings to optimize visualization.
Decreasing the saturation can help highlight the capture area.
You can change the projection type to view clusters as a UMAP
And change the view to see clusters plotted individually
You can visualize Graph-Based or K-Means clusters for both Spatial and UMAP projects.
Customize Cluster Visualization
Customize Cluster Visualization
View as many or as few clusters as you like by checking/un-checking the boxes in the sidebar. We will take a look at cluster 5.
Zoom in to see how closely clustering corresponds with the tissue morphology
Transcript Localization and Mapping Cell Types
Transcript Localization and Mapping Cell Types
We can utilize the features tab to identify cell types using genetics markers.
Let's begin by viewing marker genes for goblet cells, which can be found in normal colon mucosal tissue. We can start creating a gene list.
In the search bar type FCGBP and MUC2 to add them to our Goblet Cell marker list.
We can view each gene individually or combined.
To optimize visualization we can change the color scheme
We can also change the scale value to Linear to display UMI counts.
We can see that there are some bins scattered outside of the glandular tissue. We can filter barcodes to exclude bins with low UMI counts. If we filter barcodes by a minimum UMI count of 2, we can eliminate these bins.
Now that we have defined our goblet cells we can assign these barcodes to a cluster.
We now have a Goblet Cell cluster!
Let's identify another cell type based on a list of marker genes. This time, instead of searching for genes individually we will import a list from a csv file. Heading back to the Features tab we can select the upload option in Create a new feature list.
We will select a list of tumor markers to identify the tumor cells in our sample (this csv file is located in the same directory as the cloupe file
\\SMDNAS02.urmc-sh.rochester.edu\GRC_Workshop\Visium_HD_Colorectal_Cancer_FFPE
Or can be downloaded from the following link:
We can now see the areas that contain high levels tumor marker expression.
Let's change to the linear scale and filter barcodes.
Let's assign these barcodes to new cluster "Tumor Cells" within our CRC_Sample group.
This step can be repeated to define additional cell types.
Differential Expression and Co-expression
Differential Expression and Co-expression
We can run differential expression between our tumor cell and goblet cell clusters by returning to the Clusters tab.
Unfortunately we cannot change the order of the clusters listed in our CRC_Sample group. Since we typically want to see our differential expression results in context to our experimental group, let's quickly create a new group with the clusters listed as:
- Tumor Cells
- Goblet Cells
We will click on "Create a new group" and name it "Differential Expression".
Return to the Features tab, select the Tumor gene list, filter barcodes and save to the "Differential Expression" group.
Repeat the process with the Goblet cells.
With both the tumor and goblet cell clusters selected click on "Run Differential Expression" and then select the option to compare "Between selected cluster(s) themselves". Lastly click "Start analysis"
We can now view the differential expression results to view genes that are upregulated in the Tumor cell cluster.
If desired you can change the Feature Table Settings by clicking on the gear icon.
You also have the option to export the Feature Table.
Next we will explore the Co-expression feature in Loupe Browser. Let's say we are interested in investigating areas of glandular tissue that have been infiltrated by tumor cells. We can look for bins where tumors markers and goblet cell markers are co-expressed.
Select the Co-expression feature and choose a cell type for each axis. We can now see the areas of glandular tissue that express high levels of Tumor and Goblet cell markers. The current view can be exported as a a low resolution or high resolution image.
We can also display co-expression of specific genes. For example, say we are interested in identifying SPP1 + Macrophages. First we will need to add the individual genes in the Features tab. Click on "Create a new feature list", search for LYZ (a Macrophage marker) in the search bar and select. We can rename the list "LYZ". Repeat this process for SPP1.
As in step 19, we can now choose the individual genes for the X and Y axis
Expression Distribution
Expression Distribution
Lastly we will touch on visualizing expression distribution. In Loupe Browser we can create heat maps and violin plots that display the expression of a feature(s) across clusters.
To create a violin plot we can return to the Features tab, and create a new feature list and search for a gene of interest using the search bar. Today we will look at the expression distribution of CXCL2.
This plot is interactive and will display more detailed expression data when hovering over a cluster. This plot can also be exported as a high or low resolution image.
To create a heat map including several clusters, we will move back to the Cluster tab. With your clusters of interest selected simply click on the Heat Map button to view gene expression across these clusters. This plot is also interactive and will display the gene name when hovering over sections of the heat map.
You have the option to filter features by selecting the gear icon, and this plot can also be exported as a png or svg image.
Advanced Features
Advanced Features
Assistance with the use of the Reanalyze and Advanced Selection features will be provided during the hands-on session.