WorkflowBowBlast

WorkflowBowBlast runs automaticaly several tools to process raw Illumina reads, such as trimmomatic, bowtie2, bedtools genomecov and blastn.

1) Installation

git clone https://github.com/HeloiseMuller/WorkflowBowBlast

2) Running the workflow

Overview

It is advised to create a new directory for each project, WorkflowBowBlast will create several directories for the outputs.

To use WorkflowBowBlast, one needs to set the configuration file which contains the location of the data, the choice of the tools to run with their location and some other parameters.

Then run the main script with ./main.sh config.txt

Config file

Here is an example of config.txt

#Directory of the project
dir=/home/usr/Project
#Path of the forward raw Illumina reads
rawIllumina=/home/usr/Project/Sequencing/X22DC1_1.fq.gz

#Name of the sample
sample=data

#Reference genome of the sample
path_species=/home/usr/ref/genome1.fasta
species=speciesA

#Reference species of a second species (comment if one does not need it)
path_species2=/home/usr/ref/genome2.fasta
species2=speciesB

#Scripts to run
integrity=FALSE
trimmomatic=FALSE
bowtie2=FALSE
coverage=FALSE
blastn=TRUE
parallel=FALSE

#Number of threads to use (even number)
threads=20


#Path of software to run
path_fastqc=/opt/FastQC-0.11.8/fastqc
path_trimmomatic=/opt/Trimmomatic-0.38/trimmomatic-0.38.jar
path_bowtie2=/opt/bowtie2-2.3.4.2
path_samtools=/opt/samtools-1.9/bin/samtools
path_bedtools=/usr/bin/bedtools
path_blastn=/usr/bin/blastn 

dir=/home/user/Project is the directory of your project. All the outputs will be created in this directory.

rawIlumina=/home/usr/Project/Sequencing/X22DC1_1.fq.gz is the path of the raw Illumina forward reads. No need to specify the reverse reads, they need to be at the same location.

sample=data is the name of the sample on which one wants to run the workflow.sample cannot contain any dots.

path_species=/home/usr/ref/genome1.fasta is the path of the reference genome of the Illumina reads. The name of the species used in the outputs is $species. $species cannot contains any dots

path_species2=/home/usr/ref/genome2.fasta is the path of a reference genome of another species of interest. The name of this other species used in the outputs is $species2. $species2 cannot contains any dots. This second species should be used if one wants to look for chimeric reads between $species and $species2. If not, comment these options.

integrity, trimmomatic, bowtie2, coverage and blastn are the tools WorkflowBowBlast can run. They are detailed into more details thereafter. Set TRUE to run, FALSE to ignore them. parallele=TRUE allows WorkflowBowBlast run blastn without waiting the end of bowtie2 and coverage.

threads=20 set the total number of threads to use. If bowtie2=TRUE and blastn=TRUE, threads will be split between them if they run at the same time.

Steps of the workflow

Integrity

This option calculate the md5 of the raw sequencing reads and run fastqc

Trimmomatic

This option runs trimmomatic on the raw sequencing reads. For the following, one needs trimmed reads, so set trimmomatic=TRUE. If trimmomatic has already been ran, one can set trimmomatic=FALSE but check that the trimmed reads are located in $dir/trimmed_data/ and that they are named as follow: $sample_trimmed.1P.fq.gz, $sample_trimmed.2P.fq.gz, $sample_trimmed.1U.fq.gz and `$sample_trimmed.2U.fq.gz.

Bowtie2

This option runs bowtie2 on the trimmed reads against genome. If genome2 is specified, it also run bowtie2 on this other genome independently. After bowtie2, the resulting sam files are turn into sorted bam files and merged together.

Coverage

This option runs Bedtools genomecov on the output(s) of bowtie2. The output is a three colomns file, contening the sequencing depth at each position of the genome. Coverage also automatically calcluates the average sequencing depth over the genome and the percentage of genome covered at least once by the Illumina reads.

Blastn

This option can be ran at the same time as bowtie2 if parallele=TRUE. In this case, threads will be equally shared between blastn and bowtie2. This option firslty processes the trimmed fq.gz files to obtain one fasta file. If genome2 is specified, blastn is ran a first time on the fasta file against this genome2. Then, only reads that mapped on genome2 are mapped on genome1.If genome2 is commented, all reads of the fasta file are mapped on genome1.