BrianLandry
commented
8 years ago Toubleshooting process:
Take randomization matrix from plate 0 (not randomized), and paste in the time points for the top LED transitions for each waveform. These were then converted to Hour:Minute format. I then manually incremented the simulation time in plate 0 to identify the first on>off transition each active well undergoes, and confirmed it matched with the value for each waveform I have previously pasted in excel (this step was a positive control)
I then uploaded the savefile.irs file for both plate 1 and plate 2. I manually incremented the simulation time to identify the first on>off transition each active well undergoes. I used this to manually descramble the assignment of each waveform to a well in column "Collected Watching Plate View" where each entry is the location in format RC (#).
These were compared to the descrambled values provided from the randomizationMatrix for Plate 1 and Plate 2. Brief examination of this data suggests that there is randomization occuring, however the randomization does not seem to be accurately representing the randomization which is occurring. However this data needs to be analyzed more thoroughly. Rayka Analysis.xlsx
SynbioLucas
From Rayka's Email:
Hello Iris Developers, I think there may be a bug in somewhere in the randomization matrix.
I've attached 3 files. The 1st file (without a number) shows roughly how I programmed the plate before randomization (I X'd out the wells I didn't use before I randomized for the 2nd and 3rd files). The 2nd and 3rd files (plate 1 and plate 2) show the result of clicking 'Randomize the positions'.
When I watch the plate using plate view on Iris I can tell which well is which original well but when I look at the randomizationMatrix.csv the well numbers don't make sense. 20160401_135550_Plate.zip 20160405_091436_Plate 1.zip 20160405_091631_Plate 2.zip