Nextflow pipeline for calling SNPs based on GATK best practices

Table of contents

1. Overview
2. Installation
3. Running the pipeline
4. Dependencies
5. Credits
6. Citation

Overview

Here is a high-level summary of the pipeline:

• Convert BAM to fastq.
• Fastqc on the input files
• Trim Galore! on the input files to trim reads and repeat quality control on trimmed reads.
• Align reads to a reference genome with BWA-MEM
• Sort, index and return statistics with samtools
• Remove duplicate reads with picard
• qualimap on deduplicated reads
• Not run: GATK to realign the reads at the positions where there are indels (this was deprecated in GATK 4)
• Base recalibration with GATK tools BaseRecalibrator, ApplyBQSR and AnalyzeCovariates
• SNP calls with gatk HaplotypeCaller.
• Multiqc to summarise the various QC checks.

See main.nf for full details.

Installation

This pipeline itself needs no installation - NextFlow will automatically fetch it from GitHub rbpisupati/nf-haplocaller The pipeline runs with singularity container (based on environment.yml file included with package).

git clone https://github.com/Gregor-Mendel-Institute/nf-haplocaller.git

Running the pipeline

Basic usage

Assuming you have:

• cloned the repo into directory library
• a set of BAM files to process in directory data
• a FASTA file gibing a reference genome to align to in directory data
• a valid, active installation of NextFlow

then a minimal command to run the pipeline is:

nextflow run library/nf-haplocaller/main.nf \
    --reads "data/*bam" \
    --fasta "data/TAIR10_wholeGenome.fasta" \
    --outdir output_folder \
    -profile cbe

This will take a long time, so it is recommended to run this in a detatchable window, such as tmux.

Options

Here is a full list of arguments and options

Input

• --reads: Path to input files. This will usually include a wildcard to include all files matching a pattern, and be enclosed in double quotes ("").
• --fasta: Optional path to a reference fasta file to align reads to.
• --file_ext: File type of the input files. Options are "bam", 'fastq' and 'aligned_bam'.
• --singleEnd: Flag for whether data are single- or paired end. Defaults to false.
• -profile: Give a nextflow profile to allow the pipeline to talk to the job scheduling system on your machine. Valid inputs are:
  • mendel for PBS systems
  • cbe for SLURM systems
  • singularity
  • local to run on a local machine

Processing

• --saveTrimmed: If true, keep trimmed data. Defaults to false.
• --notrim: If true, skip trimming reads. Defaults to false

• --clip_r1 Integer number of bases to trim from the 5` end of read 1.
• --clip_r2 Integer number of bases to trim from the 5` end of read 2.
• --three_prime_clip_r1 Integer number of bases to trim from the 3` end of read 1.
• --three_prime_clip_r2 Integer number of bases to trim from the 3` end of read 2.

Output

• --project: Project name
• --outdir: Path to directory for the results.
• --cohort: Optional. Specify a group of samples to lump together into a single output file.

• --email: Optional email address to contact when the pipeline finishes.
• --plaintext_email = If true, send the notification email in plain text.
• -w: Path to working directory. Defaults to the current working directory. Note that w is preceded by only one hyphen.

Credits

• Rahul Pisupati (rahul.pisupati[at]gmi.oeaw.ac.at)
• Fernando Rabanal (fernando.rabanal@tuebingen.mpg.de)

Citation

Please cite the paper below if you use this pipeline.

Pisupati, R. et al.. Verification of Arabidopsis stock collections using SNPmatch, a tool for genotyping high-plexed samples. Nature Scientific Data 4, 170184 (2017). doi:10.1038/sdata.2017.184