Inquiring about the code for the projection between species.

[cghchuwudai]

Excellent work published in Nature Communications! After reading your article, I have some questions regarding the projection of single-cell transcriptomics between species. eg: How did you perform down-sampling? Is there code available for comparative analysis between different species? Looking forward to your response!

[DrLukeSimpson]

Hello and thank you for your comment. Regarding the downsampling, I used this code to randomly downsample: seurat_smaller <- seurat[, sample(colnames(seurat), size =25000, replace=F)]. However, Seurat have now introduced in V5 a feature called SketchData https://satijalab.org/seurat/articles/seurat5_sketch_analysis which would probably be better as it better preserves rare cell populations. Regarding the comparative analysis between species, this first requires the datasets being merged to be converted to a common nomenclature. I used the BioMart data mining tool to get lists of high confidence one to one orthologues: https://www.ensembl.org/biomart/martview/24ac7222072255a54b569b7d3edd4df6. This step may differ depending on which species you are comparing. I then convert gene names from the species of interest to a common gene nomenclature and remove low confidence, many to many and one to-many orthologues. After that, you can simply use the FindMarkers and FindConservedMarkers functions in Seurat. If you give me some more details of what you would like to do I can give you some tips. I will also add a post with an example if you like.