Skip to contents

Fit multivariate Cox regression model based on INCOMMON classes.

Source: R/cox_fit.R
cox_fit.Rd

Fit multivariate Cox regression model based on INCOMMON classes.

Usage

cox_fit(
  x,
  gene,
  tumor_type,
  survival_time,
  survival_status,
  covariates = c("age", "sex", "tmb"),
  tmb_method = "median"
)

Arguments

x

An object of class 'INCOMMON' containing the classification results as produced by function classify.

gene

The gene on which patient's stratification is based.

tumor_type

The tumor type of patients to stratify.

survival_time

The variable in clincal_data to be used as survival time.

survival_status

The variable in clincal_data to be used as survival status.

covariates

The other covariates to be used in the mutlivariate regression.

tmb_method

The method to define the reference value for tumor mutational burden TMB

Value

An object of class 'INCOMMON' containing an additional object survival.

Examples

# First load example classified data
data(MSK_classified)
# Perform Cox regression based on the classification of KRAS mutant samples of pancreatic adenocarcinoma
MSK_classified = cox_fit(x = MSK_classified, tumor_type = 'PAAD', gene = 'KRAS', survival_time = 'OS_MONTHS', survival_status = 'OS_STATUS', covariates = c('age', 'sex', 'tmb'), tmb_method = ">10")
#> ℹ There are 20289 different genotypes
#> ℹ The most abundant genotypes are:
#> • Mutant TP53 with LOH (562 Samples, Frequency 0.02)
#> • Mutant KRAS without AMP (199 Samples, Frequency 0.01)
#> • Mutant KRAS without AMP,Mutant TP53 with LOH (149 Samples, Frequency 0.01)
#> Call:
#> survival::coxph(formula = formula %>% stats::as.formula(), data = data %>% 
#>     as.data.frame())
#> 
#>                                  coef exp(coef) se(coef)      z        p
#> groupMutant KRAS with AMP     0.41896   1.52039  0.13594  3.082  0.00206
#> groupMutant KRAS without AMP  0.13670   1.14649  0.13460  1.016  0.30980
#> AGE_AT_DEATH>68              -1.32060   0.26698  0.23843 -5.539 3.05e-08
#> AGE_AT_SEQUENCING>67          1.23185   3.42756  0.23771  5.182 2.19e-07
#> SEXMale                       0.13644   1.14619  0.06167  2.213  0.02693
#> TMB_NONSYNONYMOUS> 10         0.56316   1.75621  0.26202  2.149  0.03161
#> 
#> Likelihood ratio test=70.61  on 6 df, p=3.068e-13
#> n= 1066, number of events= 1066 
#>    (656 observations deleted due to missingness)