How to do: 
Training of thresholds for iBRET method
-Start iBRET via webstart or GUI using gui.sh[Linux/Mac]/gui.bat(Windows)
-Choose method "Training" (you may also choose "Classification" at the same time if you want to perform classification after training)
-Load Data
  -Choose input file for positive reference interactions:
    -Press "Browse"
    -Browse your computer for a data file containing bret data of all positive reference interactions (e.g. using our file PRS_Data.txt (available from website or data subdirectory))
    -Press "open"
    -The name of the data file should now appear in the text field
    -Press "load" to load previously the data file
  -Repeat the last steps for negative reference interactions:
    -e.g. using our file RRS_Data.txt
-Set parameters for cross-validation to determine thresholds
  -Set number of folds: use e.g. 10, as described in online methods
  -Set abstention rate: number between 0 and 1. Sets the fraction of interactions which falls within the grey area. Use e.g. 0.25. 
  -Set repeats of CV: number wich determines, how often the cross-validation is going to be repeated. Use e.g. 2. This may take around 10 s.

Classification of protein-protein interactions measured by iBRET
-Start iBRET via webstart or GUI using gui.sh[Linux/Mac]/gui.bat(Windows)
-Choose method "Classification" (you may also choose Training at the same time if you want to perform training before classification)
-Load Data
  -Choose input file for interactions
    -Press "Browse"
    -Browse your computer for a file containing interactions you want to classify
    -Important: the data has to be in iBRET-format, see our example files PRS_Data.txt or RRS_Data.txt for more detail
    -Press "open"
    -The name of the data file should now appear in the text field
    -Press "load" to load previously added data file
    -Select which parameters you want to use for classification
  -Important: "use parameters from training" is only available if you have just trained the thresholds (see method "Training")
  -If you selected "use user-provided parameters" you are free to set your own thresholds for classification
    -Set luciferase cutoff: specify the cutoff for luciferase signals (see Online Methods for further detail). For example, you could use 15,618.
    -Set upper threshold of grey area: every BRET-ratio exceeding this threshold will be classified as positive interaction
    -Set lower threshold of grey area: every BRET-ratio below this threshold will be classified as negative interaction 
  -Results will be shown in the textfield below. 
  -To get predictions for indiviudal proteins press "save" and save the data file in any directory. 
  -Output format is: 
    -Protein 1, protein 2, BRET-ratio, classification
    -Letter code for classification:
    -P: classified as interacting
    -N: classified as non interacting
    -A: BRET-ratio is within grey area, further experiments need to be done