OD Calibrations

Experiments
1

A few questions about the OD calibrations:

  1. Are these calibration curves good / in line with the calibrations most users get? I’m getting a few r squared values of .92 and .95, and the rest close to .99. Most concerning was some RMSE values up to 0.08 . We used E.coli in PBS from ~0 to .9 OD. We did not use vial aligners, we will run another calibration using them tomorrow- this will hopefully improve our measurements.

    image
    image1920×977 233 KB

  2. If the calibration is worse than you expect, what should I do to improve it? Move the LED’s and photodiodes, or swap them out?

  3. When running the eVOLVER script on python, there’s an option that says: "calibrate vials to blank (y/n)? What does this mean?

  4. On the old tutorial for the MATLAB OD calibration script, there’s a line that says make sure LED settings are at 2125. I can’t find any settings for this in the python calibration, is this still something I have to worry about?

2

Here’s a couple screenshots of the stats on our calibration ( without the vial aligners)

image
image770×936 30.7 KB

image
image719×635 20.8 KB

3

Also some suggestions to improve the calibration process:

  1. Can you add the vial numbers to the 3D plots? There are vial number for the 2D OD 135 and OD 90 plots, but not the 3D plots
4

Hi Eric!

I’ll try to answer to your questions based on my experience.

  1. I mostly work with OD135 calibrations, so I am not sure of the quality of your data. Other users may be more insightful.

  2. When a calibration didn’t look so good, we checked that both the IR LED and the 135º photodiode were properly placed (in your case, you would also have to look at the 90º). If this didn’t improve the calibration plot, we also changed the IR LED and, if that wouldn’t work, we also changed the diode.

  3. To measure OD600 of a culture in a lab spectrophotometer, we usually place a cuvette with the media to use as a blank. The spectro (or the user) sets that absorbance as the baseline value (zero) to get an accurate measurement of the OD our cultures. When you run an eVOLVER experiment, that prompt is asking you whether you want to set the initial recorded OD as a “zero” or if you want to use the “zero” of your calibration. This is assuming that place your vials full of media but without cells, start the experiment, and inoculate your cells later on.

  4. I assume that the setting is the power of the IR LED. You can check this setting in the conf.yml file inside the eVOLVER. See here how to do it. (I assume your eVOLVER will be pre-configured and if you have a newer version of eVOLVER, the value will be different, probably 4096).

Extra 1: You can add the vial number to the 3D plots easily.

  • Go to the dpu/calibrations/calibrate.py script and look for line 173. It should look something like this:
  • After that line add: ax.set_title('Vial: ' + str(i))

I added an issue to the Github repository.

Hope it helped!

1 Like