Connect ingest and phylogenetic workflows to follow pathogen-repo-guide #19
Conversation
There was a problem hiding this comment.
Thanks @kimandrews!
I skimmed through the code changes and they look alright, but didn't review in depth. Thanks for uploading the new metadata/sequences to S3.
I ran through the phylogenetic workflow and it works as expected. The steps to upload the dataset to s3://nextstrain-data (i.e. to make it live) aren't yet here, but with the new ingest & phylogenetics being part of the master branch I think the new dataset should be made live. You could do this manually now and add the code in a later PR, but we should try to keep the pipeline in-sync with the live dataset.
Auspice flags the following error, which is not part of your PR but should be fixed at some point:
[Genome annotation] V has length 899 which is not a multiple of 3
An aside I noticed here: We should allow augur tree and augur align to run with more than one thread. You can see an example of how to connect snakemake and augur's --nthreads argument here.
kimandrews
force-pushed
the
use-new-ingest-output
branch
from
51c63ec to
5390f10
Compare
The same samples are used for example data as were previously used, but now they have been pulled from the ingest output. Ambiguous dates were manually fixed in the example data for samples JN635406.1, JN635408.1, and EF565859.1, based on dates found here: https://github.com/nextstrain/fauna/blob/dd20a1fad51433e0bdc206f4300635c0f93f8599/source-data/measles_date_fix.tsv
kimandrews
force-pushed
the
use-new-ingest-output
branch
from
5390f10 to
9e09647
Compare
Description of proposed changes
Connect ingest and phylogenetic workflows, following the pathogen-repo-guide:
This PR focuses on making changes that are necessary to move forward with creating a locus-specific workflow in a later PR.
Related issue(s)
#11
Checklist