geneSippr README

This repository contains two major scripts and one accessory script, as well as all necessary targets for GeneSipping

###modellingMultiprocessing.py

Python script used to simulate fastq files with controlled read length and fold coverage values from a reference genome. FASTQ files are mapped to gene targets with an adjustable kmer value. Creates a summary report of mapping results

###rawReadSippingUpdate.pl

Perl script used to convert bcl basecall files to FASTQ files, and subsequently map FASTQ files to user-defined targets. Creates a summary report of mapping results. Optionally can call gapGraphing.R to create histogram summary of gap size and frequency

This software was designed and run on Linux.

Requirements

The following software is necessary in order for GeneSippr to function

• Python, Perl, R
• BCL2FASTQ Conversion Software
• smalt
• samtools
• bcftools
• art_illumina - Only required for the modelling script

Use

###Installation

Install the repo using the method of your choice

For instance, in your terminal, navigate to the local directory of your choice and enter git clone https://github.com/OLC-LOC-Bioinformatics/geneSippr.git

Make sure the scripts have execute permissions

###Inputs

####modellingMultiprocessing.py

Requires four arguments:

• -i, --input: Input directory. Sets the path variable in the script
• -l, --readLength: List of read lengths to be used e.g. 18, 19, 20, 21, 22
• -f, --foldCoverage: List of fold coverage values to be used e.g. 1, 1.5, 2, 2.5, 5
• -k, --kmerLength: List of kmer lengths to be used e.g. 5, 7, 11

Additionally, the scrip requires sequence files in two locations:

• a reference genome in the "path/reference" folder. This genome must have a file extension of .fa* (.fa, .fasta, .fas, etc.)
• target files in the supplied "script path/targets" folder. The files in this folder must have a .fa extension. The folder supplied in the repository contains 19 gene targets specific to Escherichia coli, strain Sakai

####rawReadSipping.pl

Ideally, the MiseqOutput folder on the MiSeq should be mounted as a network drive. Either that, or copy the entire "MiSeqOutput/[run Name]" folder to the location of your choice.

As with modellingMultiprocessing.py, a folder of target sequences is required. This folder is supplied in the repository as "script path/target/allTargets"

Required arguments:

• -m, --miseqPath: The path of the folder that contains the run data
• -f, --folderPath: The path of the folder in which to place the output folder
• -o, --outPath: The name of the output folder
• -t, --targetPath: The path of the folder that contains the target sequences

Optional arguments:

• -r, --readLength: Length of forward reads to use
• -p, --project: Name of the project

###Outputs

####modellingMultiprocessing.py

• "path/tmp" folder with simulated FASTQ, mapped, sorted, and index BAM files, as well as VCF (variant calling format) files - these files can be deleted once you are sure you won't need them anymore
• "path/targets" folder with the targets
• "path/outputs" folder with a .csv report of the target mapping results

####rawReadSipping.pl

• "folderPath/outPath/project/query" containing the FASTQ files generated for the analyses
• "folderPath/outPath/project/results" containing the temporary files generated in the mapping portion of the script - these files can be deleted, though if an analysis need to be re-performed, these files can be reused
• "folderPath/outPath/project/reports" containing a .csv report of the target mapping results