theculliganman/cgap

cgap

##What it is: cgap is a gene extraction pipeline. It deals with large fastq files by parsing them with blast to speed bwa alignment on a reference file. This amounts to substantial time savings ##How it works:

![alt text][cgap workflow] [cgap workflow]: media/cgap_graph.png "cgap workflow" ##How to use it:

1. Update or clone clone the cgap repository.

• If you don't have cgap:

  git clone https://github.com/TheCulliganMan/cgap.git
  

• If you want to update cgap:

  cd <cgap_dir>;
  git pull;
  

3. Use our docker image with the required binaries, or compile your own software.

docker pull theculliganman/cgap:latest
docker run -itv <your directory>:/work theculliganman/cgap /bin/bash

Requirements:

• python [python version: 3.5.1, anaconda: 2.5.0]
• biopython [version: 1.66]
• samtools [version: 1.3]
• bcftools [version: 1.2]
• htslib [version: 1.3]
• bwa [version: 0.7.5-r405]
• novosort [version: 1.03.03]
• tabix [version: 0.2.5]
• blastn [version: 2.2.28]

3. Make necessary changes to the cgap config file

from os.path import abspath

CORES = int(6)

# WORK LOCATIONS
##Created in your current working directory
FASTQ_PATH = abspath("fastq_files")
HITS_PATH = abspath("hit_files")
HIT_FASTQ_PATH = abspath("hit_fastq_files")
BLAST_PATH = abspath('blast_files')
BAM_PATH = abspath('bam_files')
VCF_PATH = abspath('vcf_files')
DEPTH_PATH = abspath('depth_files')
CONSENSUS_FILES = abspath('consensus_files')
PHYLIP_FILES = abspath('phylip_files')

# PROGRAM LOCATIONS
MARK_DUPLICATES_JAR_PATH = "/bin/MarkDuplicates.jar"
SAMTOOLS_PATH = 'samtools'
BWA_PATH = 'bwa'
NOVOSORT_PATH = 'novosort'
TABIX_PATH = 'tabix'
BCFTOOLS_PATH = 'bcftools'

# QUALITY SETTINGS
## Minimum quality for bases to remain unmasked.
## Integers only!!!
MASK_MIN_QUALITY = int(20)

## Minimum depth for bases to remain unmasked.
## Integers only!!!
MASK_MIN_DEPTH = int(4)

## Minimum Blast Score #we used 50.0
MIN_BLAST_SCORE = float(50.0)

4. Run the cgap command.

python run_cgap.py \
  -refs_path <directory where fastas are located> \
  -forward <fastq_1_fw> <fastq_1_fw> <fastq_1_fw> \
  -reverse <fastq_2_rv> <fastq_2_rv> <fastq_2_rv> \
  -c 5 \
  -format_db #[optional, only if they need formatting.]

or in the form of a slurm script.

#!/bin/sh

#SBATCH --time=10:00:00          # Run time in hh:mm:ss
#SBATCH --mem=50G        # Minimum memory required per CPU (in megabytes)
#SBATCH --job-name=TortCgap
#SBATCH --ntasks=8
#SBATCH --error=/work/hdzoo/shared/cgap_bin/cgap/job.%J.err
#SBATCH --output=/work/hdzoo/shared/cgap_bin/cgap/job.%J.out

module load compiler/gcc/4.8
module load bcftools/1.3
module load blast/2.2
module load bwa/0.7
module load HTSlib/1.3
module load python/3.5
module load tabix/0.2
module load novocraft
module load samtools/1.3
module load blast-legacy
module load java/1.8

export PATH=/home/hdzoo/shared/software/bin:$PATH

python run_cgap.py \
  -refs_path galapagos_ref \
  -forward ab1tr1.fastq  ch1tr1.fastq  datr1.fastq  ep1tr1.fastq  m1tr1.fastq  p2tr1.fastq  va2tr1.fastq \
  -reverse ab1tr2.fastq  ch1tr2.fastq  datr2.fastq  ep1tr2.fastq  m1tr2.fastq  p2tr2.fastq  va2tr2.fastq \
  -c 8 -format_db;