Pipeline for somatic CNV detection from single-cell whole-genome sequencing data of non-cancerous mouse or human tissue.
Please see the files in the README folder for more detailed information on how to use this code
======================================================================================================================== INSTALLATION
-
Software Dependancies -- install these separate programs if you do not currently have them a) Bowtie 1 http://bowtie-bio.sourceforge.net/manual.shtml#obtaining-bowtie b) Samtools v0.1.19 https://sourceforge.net/projects/samtools/files/samtools/0.1.19/ c) Statsmodels (python package) For the best results, enter the following at the command line pip install --user statsmodels d) Matlab And this should be accessible at the command line Only required for the CBS step of the segment function
-
Create mapping reference(s) for mm10 and/or hg38, if you do not currently have them http://bowtie-bio.sourceforge.net/manual.shtml#the-bowtie-build-indexer I recommend follwing the strategy outlined in the following publication for the most accurate results https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5069701/
-
Obtain this repository At the command line, enter the following cd /desired/directory/for/code (ex. cd ~/GitRepos/) git clone https://github.com/suzannerohrback/somaticCNVpipeline
-
If updates have been made to the code At the command line, enter the following cd /desired/directory/for/code/somaticCNVpipeline git fetch git merge origin/master
======================================================================================================================== Command Line
The following commands assumes you have cd'd the the directory the somaticCNVcalling script is located If you have added the /path/to/somaticCNVpipeline folder to your PATH variable It is not necessary to cd to that folder You do not need to specify python when calling the function
Get a help message: python somaticCNVcalling -h python somaticCNVcalling --help python somaticCNVcalling [function] -h python somaticCNVcalling [function] --help
General usage (see function README files for further information): python somaticCNVcalling preprocess [options] python somaticCNVcalling map [options] python somaticCNVcalling count [options] python somaticCNVcalling segment [options] python somaticCNVcalling interpret [options]
See EXAMPLE_SHELL_COMMANDS.sh for actual examples
======================================================================================================================== CITATION
Feel free to use this code, but please provide credit if utilized for a publication Suzanne Rohrback el al., "Submegabase copy number variations arise during cerebral cortical neurogenesis as revealed by a single-cell sequencing approach," PNAS (2018). www.pnas.org/cgi/doi/10.1073/pnas.1812702115
======================================================================================================================== DISCLAIMER
Not all of the code in this repository will work on all systems "out of the box". That is because one purpose of this repository is to support continuity within the Jerold Chun lab. Portions of the code that are most likely to require modification or independent creation include: Mapping Bin location reference Segmentation There are multiple strategies for each of these actions, and it is beyond the scope of this repo to be universally compatible.