adapted code for pandas2.0

[gilgameshjw]

Because DataFrame.append is deprecated and replaced by concat

1. file methylprep/files/sample_sheets.py adjusted
2. file methylprep/processing/infer_channel_switch.py adjusted

There might be others .append in the code that need replacement, but these two are the changes I needed to get my job done.

[yhoogstrate]

I guess this one needs a patch too:

https://github.com/FoxoTech/methylprep/blob/master/requirements.txt

@marcmaxson @notmaurox could you look into this free bug fix/patch? Clean installs are breaking due to dependency problems w/ pandas

[notmaurox]

Hi @yhoogstrate - I appreciate your contribution. I unfortunately do not have write access and cannot merge. I know @marcmaxson is not at foxo anymore. I recommend reaching out to FOXO directly to see if they can assign someone new to maintain the repo.

[yhoogstrate]

Hi @notmaurox ,

Please acknowledge @gilgameshjw for the patch and not me. I contacted you because you were top committer, my apologies if that was not appropriate. I have reached out to FOXO, to see how they envision this. Let's wait a few days to enlist the options.

@gilgameshjw I pushed you a change in setup.py and requirements.txt so they are always in sync and explicitly depending on pandas 2.*.

kind regards,

Youri

[yhoogstrate]

@gilgameshjw @notmaurox FOXO did not even respond. I'm afraid this PR will not get accepted under this repo, so forking and version bumping would be my suggestion, but will come with the burden of maintenance... I think also the project's name should be renamed to get pip access

[IsaacPengLiu]

@gilgameshjw @notmaurox FOXO did not even respond. I'm afraid this PR will not get accepted under this repo, so forking and version bumping would be my suggestion, but will come with the burden of maintenance... I think also the project's name should be renamed to get pip access

So what is currently the best way to fit pandas 2.0 if we need to use methylprep

[yhoogstrate]

Patch it yourself I'm afraid ...

I had to do this for a project for myself too: https://github.com/yhoogstrate/pymetharray/tree/api_restructure - you can look at the changes, probably the branch itself won't work for you because I had to restructure the API considerably for the use case(s), which probably breaks with other code

Then install your pulled project folder through pip (pip install -e ...) with virtual environment

[luiscruvinel]

Are you still using methylprep after the Pandas 2.0 update? If so, have you patched it yourselves?

Alternatively, has anyone found a good solution to methylation data (.idat) pre-processing? Most of the python libraries that I found for this purpose also suffer from lack of maintenance/support and have outdated dependencies.

[IsaacPengLiu]

Are you still using methylprep after the Pandas 2.0 update? If so, have you patched it yourselves?

Alternatively, has anyone found a good solution to methylation data (.idat) pre-processing? Most of the python libraries that I found for this purpose also suffer from lack of maintenance/support and have outdated dependencies.

Here, the first answer using pip install for right version of Methylprep works out :
https://stackoverflow.com/questions/78668908/how-to-use-a-specific-version-of-methylprep-in-jupyter-notebook/78712454?noredirect=1#comment138783762_78712454