Skip to content

Account for altExp when converting SCE to AnnData #115

@allyhawkins

Description

@allyhawkins

In the function to convert from SCE to AnnData, any data stored as an altExp will not be converted. It does not appear that AnnData has much functionality for additional multi-omics data, but I did find this tutorial that we could use as a reference. Also see some additional thoughts below regarding ways to handle the multi-omics data.

The one remaining thing that you mentioned in your initial comments was the handling of alt experiments. I agree that it is correct to make that a separate issue, but I wanted to put some thoughts here while I am thinking about it. It looks to me that AnnData can't really handle those at this point except by adding the additional measures as "genes" and then annotating them in the metadata. I don't love that solution, as I think it puts an odd burden on the user to know that the matrix has other data. I think the best solution for now may be to simply output multiple AnnData objects, one for each modality. There is the MuData class which wraps up multiple AnnData objects, but I am not sure how easy the support within R will be, or whether that would really be worth the effort.

Originally posted by @jashapiro in #114 (review)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions