endrebak
If sb is interested in the why: pyranges has a method to take care of aligning chromosomes and strands. These are used for all operations (intersect, join, overlap, nearest, k_nearest...
I am getting closer :) Will hopefully have it ready by Friday next week :)
What is the reason for wanting to avoid sorting on the strand? I'd love to help you solve your problem. You could do `gr.df.sort_values(["Chromosome", "Start", "End"])`. This will give you...
I should really create a faq about this. I do not want to make backward incompatible changes to pyranges, but I might throw together pyranges2 someday. But what you are...
I do not have a clear overview of all the ramifications your suggestion might have. It might be that some code, not just mine, but anyone's, depends on the dfs...
The property-approach outlined above might be slow though. Then `gr["chr1"]` might become surprisingly slow (for something that looks like a dict lookup). And chains of commands would need to split...
This was expected. Thanks for reporting - I have been able to fix this for the NCLS, I should be able to do so for the nearest methods too :)...
Thanks for the reproducible example. I'll have a look soon.
Thanks again for reporting. I think I have a solution, but I want to sleep on it so you might not see a fix until the day after tomorrow.
I’m writing unit tests for this to be sure I get it right. Might take a while 🙈