ShinyVCFMultiSampleanalyser

This tool is aimed at helping to analyse small to medium vcf files of genotyped individuals.

If a file mapping individuals to populations is provided, most plots will be decorated with population info, and pairwise Fst as well as Faststructure can be conducted.

Multi-populations AFS spectrum analyses with moments are also available to test up to 24 demographic models using a dual-annealing optimization method.

A bunch of tools are also availbale to manually scan regions of the genome for :

• Genotype distribution over samples
• Pairewise missingness correlated to Identity by state (IBS)
• Scans of Fst, Pi and Dxy highlighting putative outliers (Fst values > a local quantile threshold)

An example of vcf file and popmap file can be found in example directory. This is a subset of data extracted from the Dryad dataset : https://doi.org/10.5061/dryad.kp11q of Ferchaud et al. 2015.

Important :

popmap file must list samples in the same order as in the vcf/bcf file !

To get the ordered list of samples in your vcf/bcf :

bcftools query -l my.vcf.gz

The popmap must be a text file of two columns separated by a tabulation. The first column is the sample names and the second is the population names.

ShinyVcfAnalyser is available as a docker image in dockerhub or by cloning this repository and building a new image.

If docker is not installed check howto do here :

https://docs.docker.com/engine/install/

Use of the image on docker Hub

Pull the last available version from Docker Hub registry :

docker pull mbbteam/shinyvcfanalyser

If you choose this solution, you have to refer to the image as mbbteam/shinyvcfanalyser in the commands that follow.

Download or unzip files to build a new image

git clone https://github.com/BELKHIR/shinyvcfanalyser

then

cd shinyvcfanalyser

Build the Docker image (to be done only once)

sudo docker build -t shinyvcfanalyser .

Now you have to bind directories from your system to directories inside the container

bind local directories to /Data and /Results in the container

e.g. I want to share my files in /home/khalid/projets/workspace/testData and get results in /tmp/Results

DOCK_VOLUME="--mount type=bind,src=/home/khalid/projets/workspace/testData,dst=/Data --mount type=bind,src=/tmp/Results,dst=/Results"

To run with binding and mapping host tcp port 9090 to port 3838 in the container (named vcfmultisample)

sudo docker run --rm -d -p 9090:3838 $DOCK_VOLUME --name vcfmultisample shinyvcfanalyser

The app is now accessible via a browser at : http://127.0.0.1:9090

The running container can be alternatively accessed in command line this way :

sudo docker exec -i -t vcfmultisample /bin/bash

This will open a bash console inside the container.
Some binaries are installed in /opt/biotools

To run a new container only in command line :

sudo docker run --rm -i -t $DOCK_VOLUME --name vcfmultisample shinyvcfanalyser /bin/bash

Generate a report from command line

cd /sagApp/

in this example we want to generate a pdf/html report for a VCF file located in /Data/snps.vcf.gz and its population info in /Data/popmap.txt

This will be run with default parameters :

Rscript -e 'rmarkdown::render("finalReport.Rmd",output_format ="pdf_document", params = list(vcf="/Data/snps.vcf.gz", popMap="/Data/popmap.txt") )'

This will generate a file named finalReport.pdf .

To get html output set output_format ="html_document" in the command line.

You can set some parameters by editing the file finalReport.Rmd and change one or several options in the params section :

params:
vcf: ""
popMap: ""
globalmaxMissing: 1
globalminMAF: 0
maxLD: 0.8
chr_geno_plot: 1
start_geno_plot: 1
maxSnp_geno_plot: 500
expectedK: 3
demoPOP1: 1
demoPOP2: 2
testModel: "IM"
optiMethod: "optimize_log"
palette: "Set3"
seed: 4321