ImmGenData: discrepancies in cell ontology mapping?

[prepiscak]

Hi,

Thanks for the great package!

The idea of using label.ont is great, however I'm wondering how was the mapping to the Cell Ontology (label.ont) done in the ImmGenData (or in any other dataset)?

For some similar/identical labels there seems to be discrepancy in the ontology label. For example, this looks fine:
B cells (B.Fo) = CL:0000843 follicular B cell
B cells (B.FO) = CL:0000843 follicular B cell

however, the following ontology labels differ:
B cells (B.FrE) = CL:0002054 Fraction E immature B cell
B cells (B.FRE) = CL:0000236 B cell

B cells (preB.FrD) = CL:0002052 Fraction D precursor B cell
B cells (preB.FRD) = CL:0000817 precursor B cell

B cells (proB.FrBC) = CL:0002400 Fraction B/C precursor B cell
B cells (proB.FRBC) = CL:0000826 pro-B cell

Note: This is just an example, but there may be other discrepancies.

Thanks

[LTLA]

I vaguely remember either me or @j-andrews7 doing it manually. Can't remember who did what.

You're welcome to submit a PR to fix any problems. It was pretty draining so attention may have slipped here and there.