search

hci-unihd / antibodies-analysis-issues

Issue tracker for problems in the antibodies analysis workflow.
0 stars 0 forks source link

Border line infected cells - 20200417_132123_311 - G03-0006 #75

Open tischi opened 4 years ago

tischi commented 4 years ago

{"plateName":"20200417_132123_311","siteName":"G03-0006","pixelLocation":[6732.656551374598,21068.444627786994,0.0],"analysisVersion":"06433667b2e6bbaa8118cc837e84abc6d0d5592a"}

@metavibor @imagirom It also becomes tricky with a hard cut off. The lower cell is a clear miss, I would say. However, the upper cell has very little virus and also responds quite little with serum binding.

Didn't we consider some border-line category which we would neither classify as infected nor control?

Screenshot

metavibor commented 4 years ago

@tischi @imagirom I think this is a good idea, we should think of implementing a third category... so beside "infected" and "control" we can have "not classified"