search

omerwe / polyfun

PolyFun (POLYgenic FUNctionally-informed fine-mapping)
MIT License
85 stars 21 forks source link

rpy2 causing issues? #153

Closed humanpaingeneticslab closed 1 year ago

humanpaingeneticslab commented 1 year ago

Installed a fresh version of polyfun yesterday, and the test_polyfun.py script would fail 1/2 way through.

That's when using rpy2 version 3.5.10.

Reverting to version 3.5.8 solved the problem. \o/

jdblischak commented 1 year ago

This is a known problem. Please see PR #149. Did you create a conda env using the latest polyfun.yml?

humanpaingeneticslab commented 1 year ago

Thanks John!

I guess you can close my issue.

We did a quick google search and we didn’t find the link below (PR #149)… already nicely documented :-)

On another note, the script that generates the 2,700 jobs still pulls stuff from the old broad URL; I couldn’t find in the polyfun project where that URL comes from, that despite I could locate the new address in compute_ldscores_from_ld.py:

UKBB_LD_URL = 'https://broad-alkesgroup-ukbb-ld.s3.amazonaws.com/UKBB_LD’

Have a nice day, Marc.

On Apr 14, 2023, at 09:31, John Blischak @.***> wrote:

This is a known problem. Please see PR #149. Did you create a conda env using the latest polyfun.yml?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.

jdblischak commented 1 year ago

We did a quick google search and we didn’t find the link below (PR #149)… already nicely documented :-)

No worries. Installation is tricky. Please try installing the conda env as below, and then let us know if you still get any installation-related errors.

mamba env create --file polyfun.yml
conda activate polyfun

On another note, the script that generates the 2,700 jobs still pulls stuff from the old broad URL

Could you please open a new Issue to report this problem? The more information you provide (line of the file that uses the old URL; example code to reproduce the error), the easier it will be to potentially fix

omerwe commented 1 year ago

@humanpaingeneticslab I'm closing the issue for now, please reopen if needed. I also just updated the file that points to the UKBB LD matrices, so hopefully that problem is resolved as well.

jdblischak commented 1 year ago

Quick update: finemapper.py can now be run with rpy2 >= 3.5.9 (#156)

humanpaingeneticslab commented 1 year ago

Thanks John!

On Apr 25, 2023, at 10:19, John Blischak @.***> wrote: Quick update: finemapper.py can now be run with rpy2 >= 3.5.9 (#156)