Skip to content
Snippets Groups Projects

Convert seurat

Merged Convert seurat
convert_seurat into main
Merged John Lafin requested to merge convert_seurat into main
Compare
and
3 files
+ 51
− 5
Preferences
File browser
Compare changes
docs/index.md
+ 4
− 2
@@ -32,7 +32,9 @@ sample sheet.
CellBender outputs several files. This workflow adds a prefix of the sample name to each output file:
- {sampleName}.h5: The cleaned cell-by-gene matrix containing __all__ droplets (both inferred cells and empty).
- {sampleName}_filtered.h5: The cleaned cell-by-gene containing only droplets inferred as cells.
- {sampleName}_seurat.h5: Same as above, compatible with Seurat for downstream analysis.
- {sampleName}_filtered.h5: The cleaned cell-by-gene matrix containing only droplets inferred as cells.
- {sampleName}_filtered_seurat.h5: Same as above, compatible with Seurat for downstream analysis.
- {sampleName}_report.html: An HTML report generated that provides diagnostic plots, an explanation of their meaning, and any warnings that arose.
- {sampleName}.pdf: Diagnostic plots without additional explanation.
- {sampleName}_metrics.csv: Contains final metrics and values.
@@ -44,7 +46,7 @@ Once the workflow is complete, users are recommended to examine the HTML report
### Note on Seurat
Seurat cannot read in .h5 files generated by CellBender natively. Please see details at [here](https://cellbender.readthedocs.io/en/latest/tutorial/index.html#open-in-seurat) for instructions on repackaging the .h5 files to a format readable by Seurat.
Seurat cannot read in .h5 files generated by CellBender natively. This workflow converts the output files to .h5 files that are readable by Seurat; these files are denoted with a "_seurat.h5" suffix. Note that these files lack metadata output by CellBender. If you want to retain this metadata, you should read the non-Seurat files using CellBender-provided functions. See [here](https://cellbender.readthedocs.io/en/latest/tutorial/index.html#use-output-count-matrix-in-downstream-analyses) for more information.
Alternatively, users can utilize the `scCustomize` [package](https://github.com/samuel-marsh/scCustomize) to read CellBender .h5 files into R, and continue their downstream workflows as usual.