search

maxplanck-ie / snakepipes

Customizable workflows based on snakemake and python for the analysis of NGS data
http://snakepipes.readthedocs.io
MIT License
374 stars 85 forks source link

genome.fa from URL #933

Closed hanrong498 closed 5 months ago

hanrong498 commented 10 months ago

Hi!

Thanks for the option to use reference genome directly from URL as input in createIndices. However, I had an issue with this:

On the first trial, I used a wrong URL and therefore the indexing steps cannot start properly. After realising this, I updated the URL link, and the createIndices.config.yaml was updated accordingly.

However, when I tried to re-submit the job, genome_fasta/genome.fa was not updated. My solution is to delete the folder genome_fasta; then the snakemake pipeline will re-create such file.

I am writing this in case anyone has the similar issues, and to report a potential bug.

katsikora commented 10 months ago

Hi @hanrong498 ,

thanks for sharing your issue. This has to do with how snakemake labels rules as completed. I'd expect the rule output (genome fasta) to be updated when you update the input (genome url). It remains to be tested if this behaviour persists when we upgrade the snakemake version. The worflow config.yaml is updated by our wrapper script, it doesn't depend on snakemake.

Best regards, Katarzyna