A pipeline which combines the advantage of transparently dealing with 10x sequencing data of Cell Ranger DNA, with the openness and flexibility of Ginkgo, used in a stand-alone fashion, in order to perform a multi-sample single-cell CNV analysis, on large-scale datasets
Requirements (tested versions in parentheses):
- Snakemake (5.4.4)
- python (3.5.3)
- gcc (5.4.0)
- R (3.5.1)
Python libraries:
- numpy
- pandas
- scipy
- seaborn
- argparse
- matplotlib
After having cloned the repositories and installed the requirements cellranger-dna and its appropriate reference genomes tarballs need to be downloaded from the 10x website (https://support.10xgenomics.com/single-cell-dna/software/downloads/latest).
A bash script to perform the whole analysis on public datasets is available:
dataset/publicdataset/cellrangerdna/sc_pipeline
Users should put the results from cellranger-dna from public datasets in the same directory before
using sc_pipeline.
For your own 10x experiments the pipeline to run cellranger-dna is in dataset/cellrangerdna
, the directory
where the fastq files can be found has to be set up in the conf.sk
file.