Entering edit mode
3.8 years ago
predeus
★
2.1k
Hello all,
It became very popular for "genomic zoologist" researchers to do single cell RNA-seq of the whole animal in order to study the available cell types. However, in almost any paper the cell type assignments are done using very few markers and by painstaking manual classification.
Do you know of any effort that would attempt automated cell type assignment for non-model organisms?
Thank you in advance, as always!
Have you looked at
singleR
(LINK)?I've seen it used in human or mouse datasets, yes. But it's really a completely different world when you go into non-mammalian cells..
Do you mean something more advanced than just using a closely related organism?
Yes - often there is no such thing as a closely related organism, since the space of scRNAseq-sequenced animals is very sparse :) but still, there must be few hundred done, and having no automated tool is quite strange.
I don't know if it's strange. Even model organisms are manually annotated (many tools exist, but they are usually not sufficient for publication-level analysis). It would be harder to do this for non-model organisms.