Plots Kaplan-meier survival curves for all the training & datasets,
independently as well as combined training datasets cohort and validation
datasets cohort. The function also plots KM survival curves for each of the
top.n.features
independently.
create.survivalplots(
data.directory = ".",
output.directory = ".",
training.datasets = NULL,
validation.datasets = NULL,
top.n.features = 25,
learning.algorithms = c("backward", "forward"),
truncate.survival = 100,
survtime.cutoffs = c(seq(5, 10, 1)),
main.title = FALSE,
KM.plotting.fun = "create.KM.plot",
plot.univariate.data = FALSE,
plot.multivariate.data = TRUE,
resolution = 100
)
Path to the directory containing datasets as specified
by training.datasets
, validation.datasets
Path to the output folder where intermediate and results files were saved
A vector containing names of training datasets
A vector containing names of validation datasets
A numeric value specifying how many top ranked features will be used for univariate survival modelling
A character vector specifying which learning algorithm to be used for model fitting and feature selection. Defaults to c('backward', 'forward'). Available options are: c('backward', 'forward', 'glm', 'randomforest')
A numeric value specifying survival truncation in years. Defaults to 100 years which effectively means no truncation
A vector containing survival cutoff time points to be used for dichotomization of patients into risk groups for senstivity analysis
A logical to specify plot's main title. Defaults to FASLE
A string containing the name of the method to use for
plotting KM curves. Defaults to create.KM.plot
Logical to indicate whether to plot univariate results for all subnetworks. Default to FALSE
Logical to indicate whether to plot multivariate results for all subnetworks. Defaults to TRUE
A numeric value specifying resolution of the png images of KM survival curves. Defaults to 100
The KM survival curves are stored under
output.directory
/graphs/
# NOT RUN {
# see package's main documentation
# }
Run the code above in your browser using DataLab