Introduction

PARSEC is a bioinformatics pipeline designed to genotype large populations using low coverage (typically <3X) sequencing data. Three imputation software are available as of today :

• Stitch
• Glimpse1
• Beagle4

The pipeline is still in development, please reach out if you need help for running it or encounter bugs

1. Index bams (SAMtools)
2. Prepare fixed size genomic chunks (bedtools)
3. Optionnal : call variants from sparse data
   1. Merge bams on each window (SAMtools)
   2. Call variants for each window (bcftools)
   3. Concatenate vcf files (bcftools)
   4. Sort vcf (bcftools)
   5. Filter variants (bcftools)
4. Impute genotypes (stitch
5. Index vcf (Tabix)
6. Concatenate vcf files (bcftools)
7. Sort vcf (bcftools)

Usage

Inputs

Aligned reads

PARSEC takes bam files as input. It is advised to perform duplicated marking and eventually BQSR recalibration on the bam files. Sarek can be used to automate read alignement (by not specifying any calling tool and eventually adding --skip_tools baserecalibrator)

Reference panel

Depending on the imputation method used, it may be necessary to supply a set of (preferentially phased) known variants (aka reference panel)

• Glimpse requires a reference panel supplied with --ref_panel
• Beagle can take a reference panel supplied with --ref_panel as a facultative input
• Stitch does not require a reference panel, but requires a set of SNP position, supplied as a vcf file (genotypes are not used). PARSEC can build it automatically using the calling subworkflow, but it is advised to validate it before running imputation. A good practice would be to run PARSEC a first time with --skip_imputation, hard filter the obtained variants, and run a second PARSEC run with --sparse_variants with the output of the first run

Tool	VCF Reference panel supplied with --ref_panel	VCF of SNPs supplied with --sparse_variants
Glimpse	✅ Mandatory	❌ Not applicable
Beagle	⚠️ Facultative	❌ Not applicable
Stitch	❌ Not applicable	✅ Recommended (hard-filtered output of a PARSEC calling run)

PARSEC main parameters

default nextflow/nf-core parameters are omitted

Define where the pipeline should find input data and save output data.

Parameter	Description	Type	Default	Required
bam	glob for input bams	string		True
ref_panel	Reference panel VCF	string
sparse_variants	VCF of variable positions used in stitch	string
imputation_tool	Imputation tool (stitch, beagl4 or glimpse)	string	stitch	True
fasta	Path to FASTA genome file. Help This parameter is mandatory if --genome is not specified. If you don't have a BWA
index available this will be generated for you automatically. Combine with --save_reference to save BWA index for future runs.	string
genome_subset	bed file specifying a genomic region to analyze	string
window_size	Size of genomic windows used for parrallelization	integer	1000000
buffer_size	Length of overlap between windows. A minimal overlap is required for good imputation	integer	100000
ngen	Stitch parameter - number of iterations	integer	1000
npop	Stitch parameter - number of haplotypes	integer	10
skip_imputation	Stop after calling/genotyping	boolean

Example usage

Note If you are new to Nextflow and nf-core, please refer to this page on how to set-up Nextflow. Make sure to test your setup with -profile test before running the workflow on actual data. ( not yet available for PARSEC)

I have a reference panel

nextflow run cguyomar/PARSEC \
   -profile <docker/singularity/.../institute> \
   --bam "/path/to/data/*.bam" \
   --fasta genome.fa \
   --ref_panel my_panel.vcf.gz \
   --imputation_tool glimpse \ # or beagle4  
   --outdir <OUTDIR>

I don't have a reference panel - Stitch

PARSEC will perform a rough SNP calling on your data (using bcftools mpileup) and then used the detected positions for stitch imputation

nextflow run cguyomar/PARSEC \
   -profile <docker/singularity/.../institute> \
   --bam "/path/to/data/*.bam" \
   --fasta genome.fa \
   --imputation_tool stitch \ # or beagle4  
   --outdir <OUTDIR>

A preferred approach is to do perform some user-defined hard filtering on the mpileup, first using option --skip_imputation, and then filtering the variant calling output (using for instance QUAL, or some external validation set).

An imputation PARSEC run can then be run using the filtered variants as a primer :

nextflow run cguyomar/PARSEC \
   -profile <docker/singularity/.../institute> \
   --bam "/path/to/data/*.bam" \
   --sparse_variants /res/of/previous/run/filtered.vcf.gz \
   --fasta genome.fa \
   --imputation_tool stitch  \
   --outdir <OUTDIR>

I don't have a reference panel - Beagle

nextflow run cguyomar/PARSEC \
   -profile <docker/singularity/.../institute> \
   --bam "/path/to/data/*.bam" \
   --fasta genome.fa \
   --imputation_tool beagle4
   --outdir <OUTDIR>

Warning: Please provide pipeline parameters via the CLI or Nextflow -params-file option. Custom config files including those provided by the -c Nextflow option can be used to provide any configuration except for parameters; see docs.

Credits

PARSEC was originally written in INRAE GenPhyse by Cervin Guyomar.

Contributions and Support

If you would like to contribute to this pipeline, please see the contributing guidelines.

Citations

An extensive list of references fornf-core pipelines bump-version 0.1.0 the tools used by the pipeline can be found in the CITATIONS.md file.

This pipeline uses code and infrastructure developed and maintained by the nf-core community, reused here under the MIT license.

The nf-core framework for community-curated bioinformatics pipelines.

Philip Ewels, Alexander Peltzer, Sven Fillinger, Harshil Patel, Johannes Alneberg, Andreas Wilm, Maxime Ulysse Garcia, Paolo Di Tommaso & Sven Nahnsen.

Nat Biotechnol. 2020 Feb 13. doi: 10.1038/s41587-020-0439-x. nf-core pipelines bump-version 0.1.0