Skip to contents

OptSurvCutR: Validated Cut-point Selection for Survival Analysis

OptSurvCutR (Optimal Survival Cut-points R) provides a rigorous, reproducible three-step workflow for discovering the optimal number and location of cut-points in time-to-event (survival) data. Designed for continuous predictors (e.g., gene expression, virome abundance, biomarkers), it moves beyond arbitrary median splits to fully data-driven stratification.

Why OptSurvCutR?

Feature Benefit
Optimal number of cuts Uses AIC, AICc, or BIC to select 0–k cut-points
Flexible search Systematic grid or genetic algorithm (rgenoud)
Covariate adjustment Control for confounders during cut-point discovery
Bootstrap validation 95% confidence intervals for cut-point stability
Publication-ready plots Kaplan–Meier, optimisation curves, Schoenfeld residual diagnostics

Installation

You can install the development version of OptSurvCutR from GitHub. Note that the genetic algorithm (method = "genetic") requires the rgenoud package, which should be installed separately from CRAN if you plan to use it.

# Core dependencies
install.packages(c("remotes", "survival"))

# Optional but highly recommended for >2 cuts
install.packages("rgenoud")

# Install the development version from GitHub
remotes::install_github("paytonyau/OptSurvCutR")

Example: Quick Workflow with CRC Virome Data 

library(OptSurvCutR)
library(survival)

data("crc_virome")

crc <- crc_virome %>%
  select(time = time_months, status, Enterovirus) %>%
  na.omit()

# Step 1: Determine optimal number of cut-points
num_cuts <- find_cutpoint_number(
  data = crc, predictor = "Enterovirus",
  outcome_time = "time", outcome_event = "status",
  max_cuts = 2, nmin = 0.15, seed = 42
)
print(num_cuts)   # BIC suggests 2 cut-points

# Step 2: Find the precise cut-point locations
cuts <- find_cutpoint(
  data = crc, predictor = "Enterovirus",
  outcome_time = "time", outcome_event = "status",
  num_cuts = 2, method = "systematic", nmin = 0.15
)

# Step 3: Validate stability with bootstrap
val <- validate_cutpoint(cuts, num_replicates = 200, seed = 456)
summary(val)

# Step 4: Visualise results
plot(cuts, type = "outcome")        # Kaplan–Meier curves
plot(cuts, type = "distribution")  # Predictor + cuts
plot_schoenfeld(cuts)               # Proportional hazards check
plot(val)                           # Bootstrap density + 95% CI

Workflow Summary

OptSurvCutR provides a three-step workflow for cut-point analysis:

  1. find_cutpoint_number(): – selects the statistically optimal number of cut-points using information criteria
  2. find_cutpoint(): locates exact cut-point values (systematic or genetic search)
  3. validate_cutpoint(): assesses stability via bootstrapping with 95% confidence intervals ## Resources

Citation 

@article{yau2025optsurvcutr,
  author  = {Yau, Payton T. O.},
  title   = {OptSurvCutR: Validated Cut-point Selection for Survival Analysis},
  year    = {2025},
  doi     = {10.1101/2025.10.08.681246},
  publisher = {Cold Spring Harbor Laboratory},
  journal = {bioRxiv},
  url     = {https://www.biorxiv.org/content/10.1101/2025.10.08.681246}
}

A JOSS submission is planned post-rOpenSci review.

Support OptSurvCutR

If OptSurvCutR helps your research, consider buying me a coffee — it directly supports ongoing maintenance with no dedicated funding.

Buy Me A Coffee

License

Licensed under the GPL-3 License.

Contact

Questions, suggestions, or issues? Please open a ticket: https://github.com/paytonyau/OptSurvCutR/issues