Data selection schemes
Data selection schemes define which data PLA 3.0 considers in the analysis. You can select, for example, which data points or dilution steps you want to include. The default scheme simply selects all data points.
About data selection schemes
When analyzing biological assays, defining which data points to include in calculations is essential for obtaining meaningful results. To support you during this process, PLA 3.0 provides data selection schemes.
Data selection schemes let you:
-
Exclude statistical outliers that may distort regression curves or calculated parameters.
-
Limit the analysis to specific dilution steps or response ranges.
-
Ensure a consistent and transparent handling of edge cases.
Dose-response analysis documents offer the following data selection scheme settings:
- Outlier detection
- Range selection
- Response range definition
Outlier detection
Use outlier detection to identify and manage response values that do not fit the expected pattern. PLA 3.0 supports replicate-based outlier detection (Dixon, Grubbs, and Standard deviation tests) and model-based outlier detection (Studentized residuals test).
How PLA 3.0 handles statistical outliers depends on the outlier processing setting you select. For example, you can exclude them from further analysis or mark them in plots and reports without excluding them from calculations.
Range selection
Range selection lets you define which dose steps of the dilution sequence PLA 3.0 includes in the analysis, thus focusing the regression on the most relevant part of the dose-response curve.
By default, PLA 3.0 uses all dose steps to fit the regression curve and calculate results. You can exclude certain dose steps, for example, if extreme concentrations lead to saturated or highly variable response values.
-
All sequence steps (default): Uses the full dose range for concentration calculation.
-
Range partial: Uses a only a range of selected steps for concentration calculation.
Response range definition
The response range defines which response values PLA 3.0 includes in the regression analysis based on their measured signal level.
By default, PLA 3.0 includes all response values. You can exclude certain values, for example, to avoid areas of the curve that are saturated or affected by background noise. To do so, define lower and upper limits for valid response values. PLA 3.0 ignores values outside of this range and does not consider them when calculating results.
Work with data selection schemes
For details on how to create and work with data selection schemes, see the Set up data selection schemes topic.