Possibility of using plinder codebase to perform splitting for other datasets

[weitse-hsu]

Hi,
Thank you for releasing this tool/dataset! The splitting algorithm presented in the paper looks quite rigorous and well-designed. However, as far as I understand, the current implementation seems to only work for the Plinder dataset. Given that the same algorithm can also be very useful for other use cases, I was wondering if it would be possible to use the existing codebase to process other datasets. If not, are there any modifications you would recommend to make this feasible?

[mbujo]

Hey,
first of all thanks a lot for contributing this rich dataset and the comprehensive documentation around it! Great work!

I would also be very much interested on running the PLINDER data pipeline on private datasets and do there similarity-based, informed and non-leaky data splits.

Could you provide any hints which scripts to execute to achieve that?

Thanks a lot!

[KefanZhan]

Hello,

I'm new to the field of protein designing, and PLINDER is the first work I read.

If I want to reproduce some work mentioned in the paper, but I don't know where to start, only to find some function files in the "/flows".

And I went across all these issues, perhaps all guys were focusing on the dataset itself. Could someone help me?

Thanks a lot!