phylo: `clean` rule removes the USVI data

[joverlee521]

Running the clean rule in the phylogenetic workflow removes the USVI data that was added in https://github.com/nextstrain/zika/pull/31

So if you run clean then try to run the workflow, it results in an error:

$ nextstrain build phylogenetic/ clean
...
$ nextstrain build phylogenetic/ 
Building DAG of jobs...
MissingInputException in rule append_usvi in file /nextstrain/build/rules/merge_sequences_usvi.smk, line 24:
Missing input files for rule append_usvi:
    output: data/sequences_all.fasta, data/metadata_all.tsv
    affected files:
        data/metadata_usvi.tsv
        data/sequences_usvi.fasta

Possible solution

I think easiest to just move the USVI data to a different directory, e.g. usvi_data, so that it does not get removed with the workflow data.

[j23414]

Other options include:

• moving USVI data back to being downloading from a URL (I think we didn't go this route because we wanted the README documentation?)
• more specifically cleaning up data/metadata.tsv, data/sequences.fasta, data/metadata_all.tsv, and data/sequences_all.fasta at https://github.com/nextstrain/zika/blob/c710ca181f23444553bceb46cb00ced9db4eece8/phylogenetic/Snakefile#L22 which would bypass creating another directory
• creating a local_data folder instead where users can spike in any of their own sequences and records