Checks the different models' accuracy against the controls and then sets the predicted values to be the best model.
1 2 3 | GetPredicted(df, Controls, Output.Dir, Target = "Assay1",
Methods = c("Linear", "Multinomial", "Manual", "SemiSupervised",
"QS.Dosage"), Console = F, LL.min = 0.5, nulliplex.thresh = 0.02)
|
df |
The dataset built from joining the results of each method specified in methods. Use |
Controls |
The Controls with Well information included |
Output.Dir |
The directory to print the resulting plots to. Uneccesary if |
Target |
The target assay. This is used primarily to rename the "Predicted" and "Probability" columns to the correct names. The assay names Assay1,Assay1#,Assay1CN, Assay1 CN, etc. will all generate a column named Assay1# and Assay1#_Pr. |
Methods |
The different methods combined with |
Console |
Should the plots be printed to the console and not saved to |
LL.min |
Any predictions with likelihood less than this will be ignored. Predictions with likelihood greater than this will be included in the final predictions weighted average. (Default=0.5) |
nulliplex.thres |
Final predictions with RQ < |
Controls |
the list of Controls for the dataset |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.