xMLfeatureplot: Function to visualise/assess features used for machine...

In Pi: Leveraging Genetic Evidence to Prioritise Drug Targets at the Gene and Pathway Level

Description

xMLfeatureplot is supposed to visualise/assess features used for machine learning. Visualisation can be made using either boxplot or dot plot for AUC and F-max. It returns an object of class "ggplot" for AUC and F-max, and an object of class "trellis" for boxplot.

Usage

xMLfeatureplot(
df_predictor,
GSP,
GSN,
displayBy = c("boxplot", "ROC", "Fmax"),
font.family = "sans",
...
)

Arguments

df_predictor	a data frame containing genes (in rows) and predictors (in columns), with their predictive scores inside it. This data frame must has gene symbols as row names
GSP	a vector containing Gold Standard Positive (GSP)
GSN	a vector containing Gold Standard Negative (GSN)
displayBy	which statistics will be used for displaying. It can be either "boxplot" for features themselves, "ROC" for AUC in ROC, "Fmax" for F-max in Precision-Recall curve)
font.family	the font family for texts
...	additional parameters. Please refer to 'lattice::bwplot' for the complete list.

Value

an object of class "ggplot" for AUC and F-max, and an object of class "trellis" for boxplot

Note

none

See Also

xPredictROCR, xPredictCompare

Examples

RData.location <- "http://galahad.well.ox.ac.uk/bigdata"
## Not run: 
gp <- xMLfeatureplot(df_predictor, GSP, GSN, displayBy="ROC")

## End(Not run)

Pi documentation built on Nov. 29, 2021, 3 p.m.