MaximilianLombardo
commented
3 years ago Hi Vinko,
Thank you for your kind words and for using cellxgene! We are excited to have you as a part of the community!
So regarding non-transcriptomic assays, cellxgene is actually quite capable of handling anything that can be put into a matrix format, so it just depends on what your feature space is (you might have features other than genes for open chromatin sequencing for instance) and if that can be put into a features x samples matrix. For bringing in flow cytometry or mass spec data, the construction of a cellxgene compatible anndata object should be relatively straightforward. Here is a link that describes the data format requirements/expectations for loading data into cellxgene. Feel free to reach out if you encounter any problems creating an anndata object with your non-transcriptomic data.
That being said, cellxgene does not currently offer some of the assay specific visualizations that you might be used to seeing with non-transcriptomic data (such as looking at a genome browser for open chromatin data or displaying vectors/predicted trajectories for velocity data). Depending on what type of assay/model you are trying trying to visulalize, it may be possible to circumvent this limitation be being clever about using the cell/sample metadata. Let us know if you have any specific needs/scenarios in mind and we can try to lend an opinion on if the data is already suitable for cellxgene.
Also, please let us know if you have specific features that you would like to see in cellxgene to accommodate non-transcriptomic data!
Cheers,
Max
Hello!
CellxGene is by far my favourite viewer and kudos for doing a great job developing it and maintaining it!
What are the odds of extending cellxgene to handle other data modalities? I am now primarily interested in flow cytometry and mass cytometry, but scRNAseq assays are expanding as well, with wealth of readouts (ADT, VDJ) and models (velocity) worth visualising.
Thanks, Vinko