search

wodanaz / Assembling_viruses

0 stars 0 forks source link

issues with pangolin upgrade #51

Closed wodanaz closed 3 years ago

wodanaz commented 3 years ago

John, I am sorry.

I need your help

I am getting the following error after I upgraded pangolin: image

how can I solve this?

Thanks

johnbradley commented 3 years ago

I think this error is coming from pangolin: https://github.com/cov-lineages/pangolin/blob/9e57a5be1c2934eb3a4be1b2fd075bed3a57d43f/pangolin/command.py#L33s Apparently the pangolin --update command didn't fully update everything it needed. You could try running conda install -c bioconda pangolin after activating escapevariants to update to the latest bioconda version. Then you should run pangolin --update again.

If this resolves your problem update the version of pangolin in environment.yml for the version of pangolin you just installed: https://github.com/wodanaz/Assembling_viruses/blob/9625b9e06588aee6d6299f2ab1f20257ec92cfd1/environment.yml#L7

wodanaz commented 3 years ago

Okay, I am working on re-activating escapevariants

johnbradley commented 3 years ago

How to activate the escapevariants conda environment from an interactive HARDAC session:

module load Anaconda3/2019.10-gcb02
conda activate escapevariants
wodanaz commented 3 years ago

Thanks John for the rapid response. I reactivated the escapevariants and seem it will work. Will let you know if something extraordinary happens.

Thanks so much for the super fast response, I greatly appreciate it

Best,

Alejo

wodanaz commented 3 years ago

It didn't work but I managed to use an alternative pangolin method to call variants. And sent the report

I put an issue in pangolin github and they recommend to do the following:

pip install git+https://github.com/cov-lineages/constellations.git

Is there any way to add this to the environment?

Thanks!

johnbradley commented 3 years ago

You can run pip after activating the conda environment to install software, there are just some gotchas to keep in mind with using pip inside a conda environment: https://wiki.duke.edu/pages/viewpage.action?spaceKey=HAR&title=Installing+custom+add-on+packages#Installingcustomaddonpackages-Usingpip

The best approach would be to update environment.yaml with the latest pangolin then and delete and recreate your environment. The latest pangolin conda environment includes constelations: https://github.com/cov-lineages/pangolin/blob/8965e57b51a4cc488c52dfbb7abd06df9bd62f60/environment.yml#L18

wodanaz commented 3 years ago

It worked by just changing the version in the initial environment set up.

cat environment.yml name: escapevariants channels:

  • bioconda
  • conda-forge dependencies:
    • seqtk=1.3
  • pangolin=3.0.3