Challenges during the upload process.

[hariszaf]

Since we started to actually use the upload process module, it is a good thing to have a place to discuss challenges people experienced when trying to upload their data.

Please feel free to share descriptions of your data and/or experimental set-ups, screenshots and anything that could help understand the problem(s) you faced.

Your input will be rather essential to us for making the upload process more flexible and μGrowthDB capable of supporting set-ups we may currently miss.

[hariszaf]

Hi there,

@PallabitaSaha had a good catch today, and we currently lack an easy way to have dead/live cell counts.
(at least in case I am not overloaded and I'm forgetting something, especially since I mix issues between CellScanner and μGrowthDB quite often 😆 )

I was thinking that it would be quite straight forward to have something like the s.t.d choice,

that when choosing it, it would add two columns
<speciesName live> and <speciesName dead>, while the original would still be there
standing for the total, in case for some reason, for a species you do not know what's dead, what's alive.

@AndrewRadev @hallucigenia-sparsa what do you think ?

[AndrewRadev]

We talked about this at some point, yeah. In terms of UI, this makes sense to me, the more interesting part is how to set this up on the database level. When I was thinking about this before, I was considering a "subtype", for a technique, so that you would have as a result 3 different techniques with separate ids:

• FC live: type = "fc", subtype = "live" (or "viable"?)
• FC dead: type = "fc", subtype = "dead"
• FC total: type = "fc", subtype = "total"

And then a MeasurementContext would be unchanged, pointing to a single technique with a techniqueId. We'd have to experiment with a specific study upload to see if this somehow interferes with comparisons or something, although it shouldn't be a problem.

We talked about one of these 3 being calculated from the others, but as you say, it's possible that you don't have live/dead breakdowns for some species or even for some particular bioreplicates, so having all 3 columns available for entry might be best. If we calculated one column based on the others, it would be a good idea to think of some UI indicator that shows that this is calculated by us and not entered by the uploaders.