Single-cell RNA-Seq Count Matrix

Main Table

The main table is divided into two independent tables for cells on the left and features on the right side (Figure 1). Both tables can be ordered and filtered separately, while the number of visible rows is shown in the top-right corner:

• Left side:

• Cell: The cell or barcode name.

• #Features: The number of features with counts (>0) for this cell.
• Counts: The sum of counts for this cell along all features.

• Right side:

• Feature ID: The unique ID of the feature or gene.

• Name: Names can be more descriptive, this column may contain duplicates.
• #Cells: The number of cells with counts (>0) for this feature.
• Counts: The sum of counts for this feature along all cells.

Figure 1. scRNA-Seq Count Matrix object main viewer.

Side Panel

The count matrix side panel shows all the available functionality for this type of data.

Actions

Click on the corresponding links to access the specific documentation for each analysis:

• Filtering. Remove low-quality cells from the count table.
• Clustering with Seurat. Group cells with similar gene expression patterns.
• Trajectory Analysis with Monocle3. Order cells in pseudotime.

Figure 2. Parts of the violin plot.

Charts

Violin plots showing the distribution of different cells' statistics. Each dot represents a cell, the position on the Y-axis is the value of the given statistic, while the position on the X-axis is arbitrary. The width of the violin (background color) shows the density of dots on each region in the Y-axis. The box plot shows the median, the interquartile range, and the outliers (Figure 2).

• Distribution of Counts: Shows one violin plot per sample with the total number of counts per cell (Figure 3). The total number of counts is computed by summing all the gene counts for a given cell.
• Distribution of Expressed Genes: Shows one violin plot per sample with the total number of expressed genes per cell, that is, genes with an expression level > 0.
• % Mitochondrial Genes Distribution: Shows one violin plot per sample with the percentage of expressed mitochondrial genes per cell. The percentage is computed taking into account the expressed genes. In order to do the calculation, a list of the mitochondrial genes has to be provided in a text file. The text file must not contain headers and one gene per line. The gene names or IDs have to be found whether in the "Feature ID" or "Name" columns (Figure 4).

Figure 3. Violin and box plots show the distribution of the total number of counts per cell.

Figure 4. % Mitochondrial Genes Distribution wizard and example file.

Export

• Export Count Matrix:Export the count matrix in Matrix Market File format. Three locations must be specified in the wizard (Figure 5) in order to save the mtx (non-zero counts coordinates), barcodes, and features files.
• Export Cell Metadata. Generate a tab-delimited text file with cell metadata. The first line is the header and one line per cell. Select on the "Subgroup(s) to Extract" which metadata from the object should be included in the file (Figure 6).

Figure 5. Export Count Matrix wizard.

Figure 6. Export Cell Metadata wizard.