Skip to content

Latest commit

 

History

History
71 lines (51 loc) · 2.27 KB

README.md

File metadata and controls

71 lines (51 loc) · 2.27 KB

FunGAP in Docker Container

This gist has instructions about runnig FunGAP pipeline from inside a Docker Container.

Requirements:

  • Docker
  • 16Gb of available disk space
  • GeneMark-ES/ET release and it's key (gmes_linux_64.tar.gz and gm_key_64.gz)

Steps

Build FunGAP docker image

Be sure you have the following files in the working directory:

Dockerfile fungap.conf gmes_linux_64.tar.gz gm_key_64.gz patch_braker.pl

GeneMark is not free for everybody, so you need to register in order to have gm_* files. If was not for that I could have push FunGAP docker image ready for use in DockerHub. The Dcoker image will have about 13Gb.

# 1. Clone FunGAP repository
git clone https://github.com/CompSynBioLab-KoreaUniv/FunGAP.git
# 2. Go to docker directory
cd FunGAP/docker
# 3. Download gmes_linux_64.tar.gz and gm_key_64.gz and put it in same directory
# 4. Build the image
docker build -t fungap .

Enter Docker image and execute FunGAP pipeline

  1. Go to the directory you have your rna-seq reads and genome fasta.

  2. Enter into a docker container of fungap:

    docker run -it -w /fungap_workspace --rm -v $(pwd):/fungap_workspace fungap bash

  3. Go to /fungap_workspace and use helper script to get Augustus species.

    python /workspace/FunGAP/get_augustus_species.py \
  --genus_name "Saccharomyces" \
  --email_address [email protected]

  4. Make protein database

    python /workspace/FunGAP/download_sister_orgs.py \
  --taxon "Saccharomyces" \
  --email_address [email protected]
zcat sister_orgs/*faa.gz > prot_db.faa

  5. Run FunGAP

    python /workspace/FunGAP/fungap.py \
  --output_dir fungap_out \
  --trans_read_1 SRR1198667_1.fastq \
  --trans_read_2 SRR1198667_2.fastq \
  --genome_assembly GCF_000146045.2_R64_genomic.fna  \
  --augustus_species saccharomyces_cerevisiae_S288C  \
  --sister_proteome prot_db.faa  \
  --num_cores 8

Now you can exit docker container. Your current working directory was mounted inside FunGAP container (on /fungap_workspace) so all output files will be available on your system.