This set of tools is designed to detect variants in a sample that has been
analyzed by amplicon-based targeted resequencing.
The three C programs (stitch, removePrimer, and qualTrim) need to be compiled.
They have been tested after compilation with gcc (version 4.8.2). To compile
with gcc, one can simply run 'make' on the command-line.
To execute the pipeline, the programs/scripts can be run via the Galaxy
platform, the command-line, or the run.sh script:
1. The included XML files wrap each of the tools for use in an instance of
Galaxy. Before running the workflow, one must compile the C programs (via
'make') and check that they and the Perl scripts are executable.
2. The command-line usage of each tool can be found by running it with '-h'
(or without any command-line arguments).
3. The run.sh script executes the entire pipeline using a pre-selected set of
parameters. It requires that bowtie2 (2.2.3), samtools (0.1.19), and VarScan
(2.3.7) be installed and available on the command-line (the path to the VarScan
jar file must be set in the run.sh script). It is run by the following command:
$ ./run.sh <FASTQ1> <FASTQ2> <BED> <GEN> <IDX> <DIR>
where
<FASTQ1> and <FASTQ2> are the paired-end sequencing files (may be
gzip compressed, with ".gz" extensions);
<BED> is the file listing the locations of the PCR primers used for
targeting, in BED format;
<GEN> is the reference genome, in FASTA format (may be gzip compressed,
but will be decompressed for the samtools mpileup step);
<IDX> is the Bowtie2-index prefix (if not available, the indexes will be
generated using <GEN> [decompressed if necessary]);
<DIR> is the directory for the output files.
Choosing different parameters, or saving various intermediate files, can be
accomplished by editing the run.sh script. Since output file names are
specified in run.sh, one should not execute multiple instances in the same
location.
- John M. Gaspar (jmgaspar@gwu.edu)
June 2015 (updated July 2016)