rakeiora-whakapapa

whakapapa workflow for rakeiora

Author: [Alastair.Lamont@otago.ac.nz] (mailto:alastair.lamont@otago.ac.nz)

Current version: 1.1

With minor folder structure adjustments to run in Rakeiora.

To use in [the Rakeiora sandbox] (https://test.rakeiora.ac.nz), define a single input of htsget_type=variants, "single" selection (it will only run on one population), restrict selection to type "population", and merge into one file whose name is input.vcf.gz (as nominated in the Snakefile).

You can make the bedfile optional in the sandbox, which means the workflow will run on the entire genome file. However, at the time of this writing, this option is not yet available in Rakeiora generally (only the sandbox). To use in the [Rakeiora analysis area] (https://analyse.rakeiora.ac.nz), you will need to provide a bedfile. Chromosome 22 bedfile is provided in the resources directory (population1.bed).

Singularity construction

In Rakeiora, the workflow will be executed in a singularity which is built as follows:

Download and extract rrBLUP package from https://cran.r-project.org/web/packages/rrBLUP/index.html (this used version 4.6.2)

Builder a docker image:

docker build -t nesi/whakapapa:1.0 - < Dockerfile

Create a singularity from this local docker image:

singularity build whakapapa_1.0.sif docker-daemon://nesi/whakapapa:1.0

Rakeiora team will make this singularity available in Rakeiora as

    /shared/singularities/whakapapa_1.0.sif

See https://test.rakeiora.ac.nz/app/sandbox/shared.xsql for info on all available Rakeiora hosted singularities.

The single rule workflow runs an R script on the multi-sample VCF file (pulled back from Rakeiora when appropriate population level visa has been granted), along with the phenotype and pedigree information (mother and father relationships).

Phenotype and pedigree Data

Initially, the phenotype and pedigree information is provided as part of the workflow (see alea/simdatafull and alea/simped103id). After ALEA is configured to contain and provide this data, it will be retrieved from ALEA. This is obviously sensitive information that will not be stored in a public repository unless it is public anonymised information, or fabricated for testing purposes.

Change log

Version 1.0 - Initial release Nov 23 2023

Version 1.1 Nov 29 2023

• move stuff to a config file (so that snakemake --lint is happy)
• loop over multiple methods (GBLUP and ssGBLUP, as configured in config.yaml)
• split stdout (output) from stderr (logs)