run_sc.md

Back to main doc.

---

Run single cell analysis

1) download files (from SODAR)

First, make sure you have the following files accessible (or download them from SODAR):

• fastq files
• ISA-tab files
• feature reference (if some samples used feature barcoding)

Note: The ISA-tab assay file should contain the columns Parameter Value[Library name mRNA] and Parameter Value[Library name sample tag]. Library name sample tag is only filled, when feature barcoding was used for a sample. Samples with and without feature barcodes can be mixed.

2) edit the config file

After creating a working directory

path/to/sea-snap working_dir

edit the config file:

vim sc_config.yaml

Set the in_path_pattern to the fastq files and the transcriptome as well as gtf files for CellRanger that are in the reference package. If the experiment used feature barcoding, set the path to the Feature Reference CSV.

3) create a sample info file

This extracts information from the fastq file paths about the samples using the in_path_pattern.

The ISA-tab assay file can also be used to extract meta-information about the samples. It should contain columns Parameter Value[Library name mRNA] and Parameter Value[Library name sample tag].

./sea-snap sample_info --from sodar --input <a_isa_assay_file> --config_files sc_config.yaml

4) run the pipeline

./sea-snap sc --slurm c

5) create a jupyter notebook from snippets

(this is under development; feel free to edit and add snippets)

./sea-snap sc l create_ipynb