Installation Using devtools/remotes

devtools::install_github("QuackenbushLab/FERRET")
library(FERRET)

You can use remotes instead of devtools because it is faster to install and run. The syntax is the following:

remotes::install_github("QuackenbushLab/FERRET")
library(FERRET)

Downloading Data

Run DownloadFERRETData() with the following parameters:

• datasetName: Dataset name (CPTAC_RCC or CPTAC_GBM)
• destinationDir: Destination directory on your machine

Evaluating Results

After running your GRN method on the data, evaluate performance using the following steps:

1. Run LoadResults() with the following parameters:
   • resultDirectory: Result directory
   • interpretationOfNegative: Interpretation of negatives ("poor" means low regulation confidence and "inhibitory" means the negative value represents an inhibitory regulation)
   • firstColumnIsRowname: Whether or not the first column is the row name (TRUE/FALSE)
   • isTabDelimited: Whether or not the file is tab-delimited (TRUE/FALSE)
2. Run BuildComparisonObject() with the following parameters:
   • sourceNetwork: The path to the file with the source network
   • ingroupToCompare: A list of file paths to use for the in-group comparison
   • outgroupToCompare: A list of file paths to use for the out-group comparison
   • results: Your FERRET_Results object generated using LoadResults()
3. Run ComputeRobustnessAUC() to obtain AUC and monotonicity scores and plot them. Include the following parameters:
   • results: Your FERRET_Results object generated using LoadResults()
   • comparisons: Your FERRET_Comparisons object generated using BuildComparisonObject()
   • metric: The metric to use for computing similarity (either jaccard, in-degree, or out-degree)
   • numberOfCutoffs: The number of cutoffs to include in the curve (default 10)
   • xlab: The label to use on the X axis of the curve
   • ylab: The label to use on the Y axis of the curve
   • mode: Either percentile if you want to compare edge weights by percentile or score if you want to compare raw scores. Default is score.
4. To write the results, run WriteRobustnessAUC() with the following parameters:
   • results: The FERRET_ROC_AUC object returned by ComputeRobustnessAUC()
   • fileName: The name of the file where you wish to save the results
5. If you want to compare multiple GRN inference methods and plot them:
   1. Follow step 3 for each relevant method / cell type / sample.
   2. Run ConsolidateRobustness() on the list of FERRET_Results objects obtained to plot the overall robustness curves.
   3. Run GetResultRanges() on the list of FERRET_Results objects obtained.
   4. Row-bind the results returned by GetResultRanges().
   5. Run MakePerformanceBarPlot() on the result from step iv.