Skip to content

zhaoxiaofei/ginkgo

 
 

Repository files navigation

This repository is a fork of robertaboukhalil/ginkgo

The core functionalities added in this fork are: (1) Support for building any user-specified genome assembly from a FASTA format file in end2end_buildGenome.sh. (2) Support for multi-processing (with mclapply) in ginkgo_parallel.sh and process_parallel.sh.

Install with conda in one command

${conda} create -y -n ${ginkgo_envname} bioconductor-ctc bioconductor-dnacopy r-devtools r-inline r-gplots r-scales r-plyr r-ggplot2 r-gridExtra r-fastcluster r-heatmap3 \
bwa bowtie parallel wgsim 
# bioconda:bwa, bioconda:bowtie, conda-forge:parallel, and bioconda:wgsim are for building genome.
# The GNU parallel greatly speeds up the genome-building process by multi-processing (running multiple jobs at once).

Build custom genome

Then, run the commands below to build your target genome:

${conda} activate ${ginkgo_envname}
cd "genomes/${GENOME_NAME_1}/original/"
bash -evx ../../../end2end_buildGenome.sh "${GENOME_FASTA_FILE}" "${GENOME_NAME_2}"
# Recommendation: keep ${GENOME_NAME_1} and ${GENOME_NAME_2} identical

Run Ginkgo to Call CNVs from scWGS data

Once the genome build is complete, you can execute Ginkgo via cli/ginkgo.sh or cli/ginkgo_parallel.sh. This implementation follows exactly the same core logic as the original robertaboukhalil/ginkgo repository.

About

Cloud-based single-cell copy-number variation analysis tool

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • PHP 35.9%
  • JavaScript 25.0%
  • R 16.4%
  • Shell 7.4%
  • C++ 6.4%
  • CSS 5.7%
  • Other 3.2%