Configure the observations
You can change the settings read from your source documents if you require.
To do so, you use the Observations editor. Usually, each row in this editor corresponds to the measurements of a single sample and dose combination. If multiple rows are created for the same sample and dose pair, the generated observations are merged. For each row, observations are generated by drawing from a normal distribution. You can configure the mean, variance, and sample size of the drawn sample.
Column | Description |
---|---|
Assay element |
Contains the name of the assay element. Upon generation, all rows with the same assay element are be assigned to a single sample. |
Dose |
Each measurement must be associated with a dose. The dose is used for calculating the mean of the normal distribution. The dose is also used to assign measurements to a specific dose step in the generated Quantitative response assays. If multiple rows with the same dose are created for a single assay element, the generated observations are assigned to the same dose step. You can use this setting, for example, to force the resulting observations to contain an outlier at a specific dose. |
Replicate |
Number of observations created for this row. |
Mean |
Mean of the observations for this sample and dose pair. Can be used as a mean for the normal distribution generating the new observations. |
Variance |
Variance of the observations for this sample and dose pair, used as the variance for the normal distribution generating the new observations. If left empty, zero is used as a variance, that is, all replicates are equal to the used mean. |
Parameters |
All other columns contain parameters and their variance. The parameters, together with the dose, can be used to calculate the mean for the normal distribution used for generating new observations. If a parameter variance is given, new curve parameters are determined for each new assay using a normal distribution. All rows of an assay element use the same parameters. Note: Depending on the model, some parameter columns
must remain empty. For example, for an assay using a 4-parameter
logistic model, only the columns A, B, C, D, and their variances
are needed. All other columns should remain empty.
|