Running without R dependencies

[afrendeiro]

Hi, could you please elaborate on what should be in the directory given as path argument when running MarcoPolo?
If I have a h5ad file produced with Scanpy could I use it as input for the algorithm and therefore skip then R dependencies?

[chanwkimlab]

Hello,

Sorry for the late reply. There was some error in my setup regarding the notification of new issues. That's a great point. I added example input files in the example directory. I also plan to have MarcoPolo directly support the AnnData object, which the scanpy package is running based on. This feature will be added to MarcoPolo in the near future.

Thank you so much for using our package! Please let me know if there are any other issues.

Best,
Chanwoo

[VPetukhov]

Would it be possible to also document what those files contain? We don't use Seurat, nor SingleCellExperiment, but from what I see it should be quite straightforward to extract this data from an arbitrary single-cell package. The only problem is knowing exactly what each file contains. It's, honestly, the only thing preventing me from trying your package :)

[chanwkimlab]

Dear @VPetukhov,

Thank you for your interest in our software. I will add a tutorial for using MarcoPolo with more detailed documentation by next week.

Best,
Chanwoo

[chanwkimlab]

Hi @afrendeiro and @VPetukhov,

Thank you for waiting. I have updated the code so that MarcoPolo can get the AnnData object as input. This will prevent many issues regarding input data preparation. I also have added docstrings and a Jupyter notebook tutorial. Using the Jupyter notebook tutorial, you can easily explore what the AnnData looks like and how MarcoPolo works. Sorry for the inconvenience caused. Please let me know if you have any issues.

Best,
Chanwoo

[VPetukhov]

Thank you, Chanwoo!