Dear organizers
I had assumed that all cytogenetic features are binary event calls, i.e., their values can only be 0 or 1.
However, my script errored out due to "CYTO_predicted_feature_XX has new levels 2", which means that there are values of 2 there.
Is this correct, and is it explained somewhere what a value "2" means in a cytogenetic feature?
Thanks!
Created by Mushthofa Mushthofa mushthofa Thank you Fadi, that's quite clear. Hi Mushthofa,
Thanks for asking. For those predicted cytogenetic features, we utilized multiple computational pipelines to make the calls. In some instances, the pipelines were able to determine that a cytogenetic feature existed in a subclonal population and those were annotated with 2's as you have seen. For a detailed review of subclonality in myeloma, please see this paper by [Bolli et al, 2014](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3905727/)
Briefly, here is our interpretation of those cytogenetic calls:
- cytogenetic call 0: we could not detect the feature in this sample
- cytogenetic call 1: we can detect the cytogenetic feature in the sample
- cytogenetic call 2: we can detect the cytogenetic feature in the sample, and we have enough depth to determine that the call is subclonal (is not coming from the main clone)
As you have seen, calls of 2 are quite rare in the dataset but sometimes can be clinically significant as discussed by Bolli et al. You can treat subclonal calls as 1's to deal with the low prevalence issue, but you lose some of the information associated with the fact that this is not a feature of the main clone detectable in the patient. I'll leave that decision up to you.
Hope this helps,
Fadi Thanks Mike! You are correct there is one CYTO feature with 0, 1, 2 or NA. I will let my colleague with more knowlege of this feature reply to this post. I'll dig a little deeper but I suspect it is a NULL vs a NA getting read in oddly. Can you point to the submission id?