pacoxph and R yield different results

Questions about ProbABEL are welcome here.
Forum rules
Please remember not to post any sensitive data on this public forum.
The first few posts of newly registered users will be moderated in order to filter out any spammers.

When get a solution to the problem you posted, please change the topic name (e.g. from "how to ..." to "[SOLVED] how to ..."). This will make it easier for the community to follow the posts yet to be attended.
jal
Posts: 5
Joined: Fri Apr 25, 2014 3:21 pm

pacoxph and R yield different results

Postby jal » Thu May 08, 2014 10:36 am

I got different answers from pacoxph and R for interaction analysis. For example, using the ProbABEL example data "coxph_data.txt", we do age*SNP interaction test. The pacoxph command is

Code: Select all

pacoxph -p coxph_data.txt -d gtdata/test.mldose -i gtdata/test.mlinfo -m gtdata/test.map -c 19 -o Ocoxph_int --interaction=2

output for first SNP:

Code: Select all

name A1 A2 Freq1 MAF Quality Rsq n Mean_predictor_allele chrom position beta_SNP_add sebeta_SNP_add beta_SNP_age sebeta_SNP_age chi2_SNP
rs7247199 GAC A 0.5847 0.415 0.9299 0.8666 200 0.333517 19 204938 0.618349 20.3844 0.0271815 0.0157229 0.0268377


Now, we combine "coxph_data.txt" and "gtdata/test.mldose" in R

Code: Select all

phe<-read.table("coxph_data.txt", header=T)
gen<-read.table("gtdata/test.mldose", header=F)
dat=cbind(phe, gen)

Let's take the first SNP rs7247199 and run the same cox mdoel in R

Code: Select all

names(dat)[9]="rs7247199"
coxph(Surv(fupt_chd ,chd) ~ age * rs7247199 + sex + othercov, data=dat)

output

Code: Select all

coxph(formula = Surv(fupt_chd, chd) ~ age * rs7247199 + sex + othercov, data = dat)

                  coef exp(coef) se(coef)      z       p
age            -0.6365  5.29e-01    1.179 -0.540 5.9e-01
rs7247199     -50.3502  1.36e-22   97.108 -0.518 6.0e-01
sex             2.3016  9.99e+00    0.402  5.728 1.0e-08
othercov        0.0923  1.10e+00    0.139  0.663 5.1e-01
age:rs7247199   0.9835  2.67e+00    1.772  0.555 5.8e-01

Likelihood ratio test=59.6  on 5 df, p=1.46e-11  n= 200, number of events= 64


You can see the results are very different. Did I do something wrong?

thanks
jal

Return to “ProbABEL”

Who is online

Users browsing this forum: No registered users and 1 guest