wasade
commented
4 years ago @adamgayoso, the support for the scipy sparse objects is identical to the data representation of biom so I agree this probably isn't too bad. Would you be interested in submitting a PR for this? I hesitate to make biom depend on AnnData, however it could be implemented as an optional dependency without too much complexity.
Hello,
I'm wondering if there is any interest in adding a feature for a conversion tool from biom data to the AnnData format. This format is used by many tools for the analysis of single-cell RNA-sequencing data, which to my understanding, can be generally applied to microbiome data (e.g., our tool for dimensionality reduction that @mortonjt made a tutorial for).
It seems like AnnData has the same general structure as biom so I don't believe it should be a difficult addition.