Imputing spatial information of scRNA genes

[michellemlin]

Hello,

I am currently working on a research project where we are looking to infer the spatial information for scRNA genes using MERFISH data. Currently, our MERFISH data is analysed using Giotto and thus, stored in a Giotto object. Our scRNA data is currently stored in a Seurat object.
I tried working with the tutorial for imputing the spatial information. However, _an issue comes up where the code is unable to work with an S4 object _.
Can you provide some guidance on how I should approach the task of integrating the MERFISH data and scRNA data if they are different types of objects?

Thank you in advance for any assistance provided.

Best,
Michelle

[Czh3]

Hi Michelle,

iSpatial is not used for imputing the spatial information of the cells derived from scRNAseq. It's for imputing the genes (which measured by scRNA-seq but not measured by MERFISH) expression of the cells derived from MERFISH. For your propose, maybe you can try cell2loc( https://github.com/BayraktarLab/cell2location/ ).

You can transform Giotto object to Seurat, then it should be easy to integrate.

Best,
Chao

[michellemlin]

Hi Chao,
Apologies, I think my description may have been a bit misleading. We are trying to achieve the same thing that you mentioned (imputing the spatial locations of scRNA-seq genes not currently in our MERFISH data).
Thank you very much for your response.

Best,
Michelle

[Czh3]

Cool, you can first transform Giotto object to Seurat, and then integrate scRNA and MERFISH to impute.

Chao