AOAC Gluten Qualitative Validation Guidance-FINAL (July 2023)

Annex D

Varying 8 Parameters in a Plackett-Burman Design

Run

Sample Size

Extraction Time

Dilution Drops

Strip Incub.

Extract Temp 13 min 13 min 13 min 7 min 13 min 7 min 7 min 13 min 7 min 7 min 7 min 7 min 10 min

Strip Load

Factor 7

Factor 8

1 2 3 4 5 6 7 8 9

1.5 0.5 0.5 1.5 0.5 0.5 0.5 1.5 1.5 1.5 0.5 1.5

3 min 3 min 1 min 1 min 3 min 1 min 1 min 1 min 3 min 3 min 3 min 1 min 2 min

60 o 40 o 60 o 40 o 40 o 60 o 40 o 40 o 40 o 60 o 60 o 60 o 50 o

7 7 3 7 3 3 7 3 3 3 7 7 5

11 11 11 11

High Low High High High Low High Low Low High Low Low Mid

High Low Low High High High Low High Low Low High Low Mid

5

11

5 5 5 5 5 8

11

10 11 12

N

1

Table 7. Result Reporting

All results from the robustness study should be reported. Below is an example result table. This is a result table from a full factorial design that varied 3 parameters, with 10 replicates per factorial pattern, but a similar table design can be used for any size full-factorial or Plackett-Burman design. The POD confidence interval (CI) should be calculated according to AOAC Appendix H.

Sample Size

Dilution Drops

Strip Incub. 7 min 7 min 7 min 7 min 13 min 13 min 13 min 13 min

Number of Positive Results

Number of Negative Results

Run

POD 0.00 0.50 0.60 0.70 0.50 0.80 0.90 1.00

1 2 3 4 5 6 7 8

0.5g 1.5g 0.5g 1.5g 0.5g 1.5g 0.5g 1.5g

3 3 7 7 3 3 7 7

0 5 6 7 5 8 9

10

5 4 3 5 2 1 0

10

Table 8. Study Analysis Both the factorial and Plackett-Burman designs can be analyzed using a regression wherein the response variable (test result) is binomial, meaning that the test result is either positive or negative. The following is an example of performing the regression on the data set in Table 8, using R and RStudio. R and RStudio are free, open access programs that can be used online, or downloaded here. Because RStudio works by writing lines of code, it is helpful to shorten the names of the column headers. When you enter the factor levels, remove any lettering or special characters, and just enter the numbers. Below is an example of how this might appear on an Excel sheet, but RStudio also allows you to upload data sets from text files and other statistics programs, or to enter the data in manually (you can learn more about using RStudio here). However you save your dataset, you want to give it a short file name as well. In this example, we are naming the Excel file “data1”.