improve testing

[hannesbecher]

Thorough testing will simplify development in the long run. It will help us to pick up mistakes sooner and make sure the pipeline runs on various datasets not only our personal favourites.

I think it would be good to (1) add a test for the pipeline to start with multiple VCF files (one per chromosome) and (2) to add the code to run this to README.md.

Specifics:

• [x] find data (per-chromosome files for the bee data should be in the shared data folder in HighlanderLab on Eddie)
• [ ] add another small test data folder to repo
• [ ] also test cluster execution
• [ ] update README.md accordingly

I know I am supposed to work on the chromosome names... coming soon.

[janaobsteter]

@hannesbecher , there is already additional files there. The previous test (when we didn't have ancestral inference in the pipelien) started with split and combined files. BUT the problem is that the ancestral allele inference is not adopted to start from multiple chromosomes. We discussed that we will move the ancestral inference to after the merge (I've already started to work on this).

However, if you already have ancestral alleles file, you can test this on those files are here: "HighlanderLab/share/Snakemake/Data/RawVCF/TestChromosomeData"

[janaobsteter]

As mentioned, the ancestral allele inference will move as planned in #61

[hannesbecher]

Thanks for clarifying!