How to create Survival Formula for Phenotype Data?
0
0
Entering edit mode
17 months ago
dec986 ▴ 300

Hello,

I'm using survminer to try to create a survival formula for a phenotype data set.

library("survival", "survminer")
data <- read.csv(file = "Study.cleaned.csv", header = TRUE)
basic.cox <- coxph(formula = Surv(Observation_time, Event_Detected) ~ Patient_Age + Sex+BMI, data = data)
basic.assump <- cox.zph(basic.cox)
sink('/home/varieties/variables.tsv')
print(basic.assump)
sink()

This produces a file that looks like this:

     rho  chisq      p
A  0.00477 0.0987 0.7534
B -0.02352 2.5383 0.1111
C  0.02513 3.2104 0.0732
GLOBAL                             NA 5.6162 0.1319

I have tried about 2,000 different combinations of each factor A through L by summing them , e.g. A+B+C, A+B+D+E... etc. but the problem is that none of the summaries have all p-values being < 0.05, which is a requirement of the study.

Also, some times factors will split, and I don't know why. The output file will look like this:

C<100       0.207766 9.59e-09 0.999922
C<50        0.188145 2.07e-08 0.999885
C<69        0.124370 9.49e-09 0.999922
C>325       0.138520 1.05e-08 0.999918
C100        0.004437 9.01e-02 0.764040
C100.0      0.189391 1.14e-08 0.999915
C101       -0.012151 6.76e-01 0.410877...

and I don't know why it's splitting like this.

I have two questions: 1. How can I determine a formula that will have all p-values < 0.05? 2. Why do some factors split like C did in the example above?

survival • 322 views
ADD COMMENT

Login before adding your answer.

Traffic: 2511 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6