AxelitoMartin/OncoCast: Ensemble learning survival prediction with improved interface.

OncoCast is an ensemble learner relying on cross-validation for high dimensional survival data. The algorithm can handle left truncated (delayed entry) data for all available methods. Multiple regression methods can be used including LASSO, elastic-net or generalized boosted regression. Originally developed to stratify cancer patients based on their genomic profile. Through cross-validation we generate an average predictive risk score that acts as a surrogate for the aggregated effects of the profile of each patients. We include an interactive R based application that enables the user to explore the results in various ways. More details can be found in the vignette included in the package.

install.packages("remotes")
remotes::install_github("AxelitoMartin/OncoCast")