[R] Cox Proportional Models and Haplotypes

[Brad McNeney]

You could do a weighted cox ph model, with possible haplotype
configurations for each subject weighted by their posterior
probabilities given genotype data.

Are your markers SNPs?  If so you can use a utility function from the
hapassoc package to get started. For example, if your data is in a
dataframe dat, with nsnp SNPs in the last nsnps columns, you could
create an augmented data frame (augmented by pseudo-individuals for
each subject with ambiguous phase) with

library(hapassoc)
ph<-pre.hapassoc(dat,nsnps)
augdat<-cbind(ph$nonHaploDM,ph$haploDM)
wts<-ph$wt

and then use coxph with augdat as the data frme and  wts as the
weights. See ?pre.hapassoc for details on accepted formats for the SNP
data.

Brad
--
Brad McNeney
Statistics and Actuarial Science
Simon Fraser University

On Tue, Oct 26, 2010 at 12:01 PM, David Winsemius
<dwinsemius at comcast.net> wrote:
>
> On Oct 26, 2010, at 8:57 AM, sr500 wrote:
>
>>
>> Hello,
>>
>> I was wondering if anyone knew of a function that fits haplotype data into
>> a
>> cox proportional hazard model. I have computed my Haplotype frequencies
>> using the haplo.stats package. I have also been using the haplo.glm
>> function
>> but this is a linear regression and is not quite what I am looking for…
>>
>
> I think you need to describe your data situation more completely and ideally
> would include a small extract of your data to allow testing and
> illustration. Cox proportional hazards models are typically used to analyze
> time to event data and you have not alluded to any "events".
>
> --
> David.
>
>> Thank you very much,
>> SR
>>
>> --
>> View this message in context:
>> http://r.789695.n4.nabble.com/Cox-Proportional-Models-and-Haplotypes-tp3013989p3013989.html
>> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>