How to re-create a DoE for bioassay qualification from a publication?

[#1]

Hi All

I'd like to re-create the following DoE for cell-based bioassay qualification and robustness testing published by  Coffey et al, 2013 (BioProcess International 11(6) June 2013). 

Objectives of qualification study: to estimate assay accuracy, intermediate precision, and linearity across the normal operating range of assay conditions; to assess robustness of relative potency and 4PL model parameters to critical assay parameters.

DoE: 2(5–2) fractional factorial with 8 replicated center points:

- 16 experimental runs are split into two blocks of eight conditions for each analyst

- two analysts running four experimental conditions on each of two days

- two test samples on each plate, alternating placement of different level potencies on and across plates.

I'd appreciate any support. 

[#2 · Accepted solution]

Hi @AgnieszkaL,

I was able to replicate a similar design for the five factors and including Day and Analyst. It won't be exactly the same as the example due to randomization.

I'm not sure how they included the information about the independent replicates for each nominal potency. Do you know whether this was also via DoE or was it something else?

Anyway, I believe this was an initial 8-run custom design that was augmented with 8 additional center points. First, I created the design via the custom design with the following factors: 

The table below is what you get. Since Day was a blocking factor, we wouldn't be able to augment the design with center points. I had to change the design role for Day by clicking the asterisk and changing it into a categorical factor: 

After this, go to DoE > Augment Design > Fill in the Y and all the factors as the X. Then, Add Centerpoints > 8 and click Make Table to get the following data:

You can then change the -1 to Low, 0 to Mid and +1 to High.

Let me know if this helps!

[#3]

Excellent! Thank you Owen - super useful. I also thought that the center points must have been augmented to the first design. The remining question is about the replicates. No details are provided in the publication. Is there a way to add this information as possibly not part of DoE but as simply a part of study information? This is needed to perform the study data analysis as follows:

[#4 · Accepted solution]

Hi @AgnieszkaL 

You could add this information manually by adding one column called Nominal Potency. For each replicate, you could copy the row within the data table.

If we were to take the first row in the publication as an example, this would be repeated 4 times in the data table.Two will have Nominal Potency of 141%, one at 200% and the last at 71%.