Downloading data from Single Cell Lands
There are multiple ways to export data after browsing Single Cell Land data.
If you are interested in the metadata for selected projects, samples, or cell clusters, you can simply click the Open in Excel or Open as Text buttons.
For most Views, to export the data used for the currently plotted View, click Open in Editor or Open in Excel in the Task tab.
In most cases, once you have identified a project that has interesting data (samples or cell types of interest, or interesting expression of your favorite genes), you will want to download the full project with metadata, expression counts, differential expression, and marker genes used by the curators.
Download Full Projects
Warning: Single Cell projects can be quite large (hundreds of MB to several GB), and Download Project will download the project to your Server Projects folder, so you should only keep downloaded projects while you need them to extract relevant data, then delete the project from your folder.
From any View within Single Cell Lands, select data from a project of interest, so the Selection Details are displayed. Then click Project Details to display the project-level metadata for your project of interest.
After double-checking that the selected project is interesting, click on the ProjectName in the Selection Details, then click Download Project in the Action section at the bottom-left.
You will be prompted with the details of the project, and you can select to download. If there are multiple projects available under that ProjectName (e.g. multiple platforms) you can use the dropdown dialog to select the project of interest.
After downloading, switch to the Analysis tab. You should find multiple objects that contain different data.
In the -Omic Data section, you will find the expression matrix for all cells.
- Raw counts data will found in the object with the "UMI" or "Raw" (nonUMI) suffix
- Raw counts are generally preferred input for downstream analysis
- Normalized TPM or RPM counts will be found in the object with the "Rpm" (UMI) or "Tpm" (nonUMI) suffix
- Normalized counts are generally preferred for visualizations, and will reflect what is displayed in the Land visualizations
In the Table section, you will find the analysis results, including * Inference report, * Dimension Reduction CellMap results, * Top Gene Markers, * metadata at the * Project * Sample * CellMap * Cluster * Cell Level.