robert-koch-institut / SARS-CoV-2-Sequenzdaten_aus_Deutschland

Ein zentraler Bestandteil einer erfolgreichen Erregersurveillance ist das Verständnis der Verbreitung eines Erregers sowie seiner pathogenen Eigenschaften. Hierbei stellt das Wissen über das Erregergenom eine wichtige Informationsquelle dar. So erlaubt der Nachweis von Mutationen im Genom eines Erregers, Verwandtschaftsbeziehungen zu rekonstruie...
https://robert-koch-institut.github.io/SARS-CoV-2-Sequenzdaten_aus_Deutschland/
Creative Commons Attribution 4.0 International
67 stars 7 forks source link

Update pangolin-data to v1.13 (it's at v1.12 right now) #31

Closed corneliusroemer closed 2 years ago

corneliusroemer commented 2 years ago

It would be great if you could update pangolin-data to the latest version, it's running at v1.12 right now but v1.13 has been out for more than a week now.

This is the current pangolin output in this repo, note the v1.12

image

These are the lineages you're missing, as they were released in v1.13

image

Here's the release: https://github.com/cov-lineages/pangolin-data/releases/tag/v1.13

mg14 commented 2 years ago

This seems relevant as BF.7 (and other BA.5* variants with S:346T) seem to be rising, but are missed by v.1.12:

image image
matthuska commented 2 years ago

Hi @corneliusroemer , we use the conda packages for pangolin and its dependencies. Currently the current conda package for pangolin-data says it is 1.13, but pangolin --all-versions reports 1.12:

$ conda list | grep pangolin-data
pangolin-data             1.13               pyh5e36f6f_0    bioconda
$ pangolin --all-versions | grep pangolin-data
pangolin-data: 1.12

This is a known issue and has been reported to the pangolin-data issue tracker: https://github.com/cov-lineages/pangolin-data/issues/22#issuecomment-1214759528

As soon as that is resolved and an updated conda package for pangolin-data is released we will automatically start using it.

mg14 commented 2 years ago

Hi Matt I think this is fixed now, see

https://github.com/cov-lineages/pangolin-data/issues/22#issuecomment-1220661321

matthuska commented 2 years ago

Yes I noticed, thanks! The new version should be automatically used in tonight's batch run and the upload to GitHub tomorrow should reflect that. I'll check tomorrow to be sure, and resolve any issues if it's not working as expected.

Thanks again, the online SC2/open-source/conda communities are amazing :)

corneliusroemer commented 2 years ago

Thanks @matthuska for explaining!

I think for the future: you can update the data yourself simply by running pangolin --update-data - not relying on bioconda to update the data first.

matthuska commented 2 years ago

Thanks @corneliusroemer , there are some technical reasons why we have to stick to conda for all updates. It can lead to delayed updates but we have some internal processes that require us to be able to exactly recreate older pangolin environments, and for that we rely on conda. Thanks for your patience.

mg14 commented 2 years ago

Hi Matt - problem solved, thanks!

https://mobile.twitter.com/MoritzGerstung/status/1560864142952730624