ALDex2 issue

[SRuwi]

Using the following command
daa_results_df <- pathway_daa(abundance = kegg_abundance, metadata = metadata, group = "Stage", daa_method = "ALDEx2", select = NULL, reference = NULL)

i am getting the following error
Running ALDEx2 with multiple groups. This might take some time, please wait patiently...
Error in Math.factor(c(3L, 3L, 3L, 4L, 4L, 4L, 1L, 1L, 1L, 2L, 2L, 2L)) :
‘round’ not meaningful for factors

The metadata file is below

and the top part of the kotoKegg file is below

I converted the second column of the metadata file (Stage) to factors as well and am getting the same error. Has anyone encountered this or have a solution? Thank you in advance

[cafferychen777]

Dear @SRuwi,

It appears that there have been recent updates to the ALDEx2 package which might have introduced some compatibility issues with ggpicrust2. For the time being, I would recommend using an alternative method to proceed with your analysis. I will investigate this issue promptly and aim to resolve any conflicts or bugs arising from the update. Thank you for bringing this to my attention, and I appreciate your patience as I work to address this matter.

Best regards,

Chen YANG

[SRuwi]

Hi Chen Thank you for responding, I appreciate this. I will look at alternatives in the mean time but ill keep and eye out for any fix that you make. Thank you. Stephen

…

________________________________ From: Caffery Yang ***@***.***> Sent: Wednesday, November 8, 2023 7:29 AM To: cafferychen777/ggpicrust2 ***@***.***> Cc: stephen.ramnarine ***@***.***>; Mention ***@***.***> Subject: Re: [cafferychen777/ggpicrust2] ALDex2 issue (Issue #67) Dear @SRuwi<https://github.com/SRuwi>, It appears that there have been recent updates to the ALDEx2 package which might have introduced some compatibility issues with ggpicrust2. For the time being, I would recommend using an alternative method to proceed with your analysis. I will investigate this issue promptly and aim to resolve any conflicts or bugs arising from the update. Thank you for bringing this to my attention, and I appreciate your patience as I work to address this matter. Best regards, Chen YANG — Reply to this email directly, view it on GitHub<#67 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASKKR7LLMXEQRQMDZ4CK27TYDNULNAVCNFSM6AAAAAA7B3QYVKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMBRG4YDENJZG4>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[zhonglinma]

Hi Chen @cafferychen777, I also got this error, is there any way to deal with it?

[cafferychen777]

Hello @zhonglinma,

For now, please change to another daa_method such as LinDA.

Best regards,
Chen YANG

[asgarhussain]

Hello @cafferychen777 , @zhonglinma and @SRuwi,
I also faced the same error while using ALDEx2 method. I am able to resolve the issue by installing ALDEx2 v.1.28 via source file downloaded from Bioconductor archive:
https://mghp.osn.xsede.org/bir190004-bucket01/index.html#archive.bioconductor.org/packages/3.15/bioc/src/contrib/Archive/ALDEx2/

I removed the currently installed version using : remove.packages("ALDEx2")
and installed using install.packages("archive.bioconductor.org_packages_3.15_bioc_src_contrib_Archive_ALDEx2_ALDEx2_1.28.0.tar.gz", repos = NULL, type="source")

[cafferychen777]

Dear @asgarhussain, @zhonglinma, and @SRuwi,

Thank you all for your valuable feedback and for sharing your experiences with the ALDEx2 issue in the pathway_daa function of ggpicrust2. I appreciate @asgarhussain's suggestion to resolve the issue by installing an older version of ALDEx2 (v.1.28) from the Bioconductor archive.

Based on your inputs, I will update the ggpicrust2 documentation to include this workaround for users encountering the same error. Additionally, I will continue to investigate the compatibility issue with the latest version of ALDEx2 to provide a more permanent solution in future updates of ggpicrust2.

Thank you once again for your contributions to improving ggpicrust2. Your patience and support are greatly appreciated.

Best regards,
Chen YANG

[Lacona]

Unfortunately I encounter the same error running ALDEx2. I installed ALDEx2 vs 1.28.0 to work around but I still get the error:
Running ALDEx2 with multiple groups. This might take some time, please wait patiently...
Error in Math.factor(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, :
‘round’ not meaningful for factors
Is there any other solution for this problem?

[cafferychen777]

Hi @Lacona,

Thank you for reaching out and providing detailed information about the issue you're encountering with ALDEx2. It seems like several users have faced similar problems.

Could you please check your sessioninfo to ensure that you're using the correct version of ALDEx2? Sometimes, users inadvertently have an older version installed, which might lead to such errors even after attempting to update.

If you're still experiencing difficulties after confirming your ALDEx2 version, please let me know, and we'll explore other potential solutions together.

Best regards,

Chen YANG

[Lacona]

Thank you for your quick reply!

The error occurs with ALDEx2_1.34.0 but also with ALDEx2_1.28.0.
Check was done running sessionInfo().

[marwa38]

ALDEx2 v.1.28 worked!

[cafferychen777]

Hi everyone,

After investigating the ALDEx2 source code, I believe I've found the root cause of the "round not meaningful for factors" error. The issue occurs when the group variable is stored as a factor type in R, which conflicts with ALDEx2's internal type conversion.

I've pushed a fix to resolve this issue. You can now get it working by:

1. Reinstalling the package from GitHub
2. Making sure your group variables are properly handled as characters before passing them to the ALDEx2 function

The underlying problem was related to automatic type conversion that ALDEx2 performs when dealing with factor variables. The fix ensures proper type handling during the analysis pipeline.

Please let me know if you encounter any other issues after updating the package.

Best regards

[marwa38]

Thanks for your help.
I haven't tried yet but mine were not factors but characters. I changed to factors at the time I was working with this and I got the same round not meangiful error for characters as well as factors.