warpedlmm installation

[asifzubair]

Hi, in the documentation there is a suggestion to use warpedlmm to transform continous variables:

This version of FaST-LMM is designed for use with randomly ascertained data with Gaussian residuals. If you have case-control data with substantial ascertainment bias, you should first transform your phenotype(s) using LEAP (Weissbrod et al., arXiv 2014). If you are analyzing continuous phenotypes with non-Gaussian residuals, you should first transform your phenotype(s) using Warped-LMM (Fusi et al., Nature Commun 2014).

I installed warpedlmm but I get the following error when I import it:

>>> import warpedlmm
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/opt/anaconda3/envs/warpedlmm/lib/python3.9/site-packages/warpedlmm/__init__.py", line 18, in <module>
    import testing
ModuleNotFoundError: No module named 'testing'

Any comments on how to fix this ? Thanks!

[CarlKCarlK]

Greetings,

We don't think that the authors of Warped-LMM are supporting it anymore. However ...

1. Last month, a Fast-LMM user in Australia said he got Warped-LMM working for himself in Python 3. If you'd like me to introduce to you him, send me a message (carlk at msn.com).

2. Also, last month, FaST-LMM collaborator, David Heckerman, asked the Australian researcher if we could see the marginal distribution of his phenotype (to see if discretization would be practical). If you like, send a plot of your phenotype distribution via email and I'll "forward all" it to David.

   • Carl From: Asif Zubair @.> Sent: Sunday, April 28, 2024 9:30 PM To: fastlmm/FaST-LMM @.> Cc: Subscribed @.***> Subject: [fastlmm/FaST-LMM] warpedlmm installation (Issue #42) Importance: High

Hi, in the documentation there is a suggestion to use warpedlmm to transform continous variables:

This version of FaST-LMM is designed for use with randomly ascertained data with Gaussian residuals. If you have case-control data with substantial ascertainment bias, you should first transform your phenotype(s) using LEAPhttps://github.com/omerwe/LEAP (Weissbrod et al., arXiv 2014). If you are analyzing continuous phenotypes with non-Gaussian residuals, you should first transform your phenotype(s) using Warped-LMMhttps://github.com/PMBio/warpedLMM (Fusi et al., Nature Commun 2014).

I installed warpedlmm but I get the following error when I import it:

import warpedlmm

Traceback (most recent call last):

File "", line 1, in

File "/opt/anaconda3/envs/warpedlmm/lib/python3.9/site-packages/warpedlmm/init.py", line 18, in

import testing

ModuleNotFoundError: No module named 'testing'

Any comments on how to fix this ? Thanks!

- Reply to this email directly, view it on GitHubhttps://github.com/fastlmm/FaST-LMM/issues/42, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ABR65P7Y3AYJJ2USL5LUT6LY7XEC7AVCNFSM6AAAAABG5TIS6GVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI3DQMBZG4YDKOA. You are receiving this because you are subscribed to this thread.Message ID: @.**@.>>

[asifzubair]

Thank you, @CarlKCarlK , I'll shoot you an email. Yes, I also tried using 2to3 to get this to work, but with little luck. I'd appreciate if the Australian user could share his repo. Thanks!