PrognosticModel: Prognostic Modeling for Survival Data

In IOBR/IOBR: Immune Oncology Biological Research

Prognostic Modeling for Survival Data

Description

This function prepares the data, splits it into training and testing sets, and then fits LASSO and Ridge regression models using the 'glmnet' package. It evaluates these models by calculating prognostic results and optionally plots ROC curves for each model. It is specifically tailored for survival analysis.

Usage

PrognosticModel(
  x,
  y,
  scale = FALSE,
  seed = 123456,
  train_ratio = 0.7,
  nfold = 10,
  plot = TRUE,
  palette = "jama",
  cols = NULL
)

Arguments

x	A matrix or data frame of predictor variables.
y	A vector or data frame of survival outcomes, typically including time and status
scale	Logical, if TRUE, scales the predictor variables.
seed	Integer, a seed for random number generation to ensure reproducibility.
train_ratio	Numeric, the proportion of the data to use for training, e.g., 0.7 for 70 percent
nfold	Integer, the number of folds for cross-validation in model fitting.
plot	Logical, if TRUE, plots the ROC curves for the LASSO and Ridge models.
palette	Optional, a string specifying the color palette. Default is "jama", which is applied if 'cols' is NULL.
cols	Optional, a vector of colors for the ROC curves. If NULL, default color palettes are applied based on the 'palette' parameter.

Value

A list containing the results from the LASSO and Ridge regression models, the training data, and optionally, the combined plot of ROC curves if 'plot' is TRUE.

Examples

data("imvigor210_sig", package = "IOBR")
data("imvigor210_pdata",package = "IOBR")
pdata_prog <- imvigor210_pdata %>% dplyr::select(ID, OS_days, OS_status) %>% mutate(OS_days = as.numeric(.$OS_days)) %>% mutate(OS_status = as.numeric(.$OS_status))
prognostic_result <- PrognosticModel(x = imvigor210_sig, y = pdata_prog, scale = T, seed  = 123456, train_ratio = 0.7, nfold = 10, plot = T)

IOBR/IOBR documentation built on April 3, 2025, 2:19 p.m.