Estimated growth curve discrepancy

Software
1

Hello,

I have recently run some simple batch cultures on the pioreactor and I’ve observed that the estimated maximum growth rate differs from what I have estimated based on manual OD600 measurements (which also tends to agree with published estimates). It looks like the pioreactor growth rate estimates are less than half of the ones I get.

Is this due to the OD900 that the pioreactor uses or a different way to estimate the growth rate?

Thank you,

2

Hi @samunozg,

the estimated maximum growth rate

Are you referring to our calculation (“growth rate” in the overview), or are you manually pulling the data to calculate this. If the former, for fast growing cultures, you might be underestimating the max growth rate if the parameters are not optimal. Try the following in your configuration for your next experiment

[growth_rate_kalman]
# obs_std ↑ smooths growth rate, rate_std ↑ more responsive growth rate
obs_std=1.5
od_std=0.0025
rate_std=0.25

These parameters are more responsive than the default ones.

If you can share an export of your raw od_readings (email is fine), I can run it through a demo growth-rate calculator with these new parameters.

I don’t think the OD900 matters much to growth rate calculations.

1 Like
3

We do use a non-standard way to estimate growth rate. We use a Kalman filter, vs the standard way of fitting an exponential to the OD / linear model of the log-OD. The differences should be minimal, since the Kalman Filter internally has the same exponential model as the standard way.

4

Hi Cam,

I agree with you. There should be little to no differences. Let me a manual growth curve myself and I’ll get back to you.

Thanks!

5

Cam, I just wanted to give you an update on this. We did a few more tests and it looks like the problem was mostly on our side. While we still see some differences, these are not nearly as large as mentioned previously. Thanks!