This function fits a per-feature survival model to the full (non-resampled) data matrix using the observed sample metadata, producing the final statistics used for ranking features.
fit_survival(x, meta.info, formula.str, competing_risks)A data matrix where rows represent features (e.g., proteins, metabolites) and columns represent samples.
A data frame containing the metadata for the samples.
Must include time, event, and any additional covariates
used in formula.str.
A string specifying the formula to be used in model
fitting. Must include a Surv(time, event) term. The per-feature
coefficient term (y) is prepended automatically.
Logical. If FALSE (default), a Cox
proportional hazards model is fitted per feature using coxph.
If TRUE, a competing risks model is fitted per
feature using crr from the cmprsk package.
A list containing the following elements:
A numeric vector of absolute coefficients (\(|\beta|\)) for each feature.
A numeric vector of standard errors of the coefficients for each feature.
A numeric vector of exponentiated coefficients (hazard ratios, \(e^{\beta}\)) for each feature.
For each feature (row), the function appends the feature expression values
as y to the sample metadata and fits either a Cox proportional
hazards model (competing_risks = FALSE) or a
subdistribution hazard model (competing_risks = TRUE). The
coefficient, its standard error, and the exponentiated coefficient
(hazard ratio) for the feature term y are extracted.
Unlike permutating_survival and bootstrap_survival, this
function operates on the observed (non-permuted, non-resampled) data to
produce the final statistics used for feature ranking.