discordance between BestCall and probabilities

[LongpanUPC]

Hi,
a very good tool helps us classify tumors into different immune subtypes!
I run this function and get the results. in the results table, I found there are some discordances between BestCall and probabilities of belonging to each subtype. For instance, in this sample, bestcall indicates C3 subclass but C4 has higher probability. So I don't know which result I need to keep?

and here is the summary infomation in the series. 18 samples have these inconsistent information

Thanks in advance!

[Gibbsdavidl]

Hi there,

Thanks for writing. The "best call" comes from a classifier that's trained on the C1-C6 scores, and sometimes doesn't agree with the maximum of the scores. There's a figure describing the classifier here (figure 2): https://www.biorxiv.org/content/10.1101/2020.01.17.910950v1

I would stay with BestCall.
-dave

[LongpanUPC]

Hi there,

Thanks for writing. The "best call" comes from a classifier that's trained on the C1-C6 scores, and sometimes doesn't agree with the maximum of the scores. There's a figure describing the classifier here (figure 2): https://www.biorxiv.org/content/10.1101/2020.01.17.910950v1

I would stay with BestCall. -dave

thank you very much!