Training classifier

[Piyal saha]

Hello,

I am trying to classify lung macrophages with cell type annotation. I have my own list of markers to classify. But I am confused about one thing. To train my classifier, do I need to use my own scRNAseq data? Or do I need to use other published data which have already been annotated with different types of cell.

Thanks in advance.

[hpl...@gmail.com]

Hello,

You can do either, either train on published data (and in that case I would recommend using the marker-free training described here https://cole-trapnell-lab.github.io/garnett/docs_m3/#1c-train-a-marker-free-classifier) or you can train on your own data using your list of markers.

Best,

Hannah