See the "newdata" argument in ?survfit.

You can use it like this:
cox.surv <- survfit(cox.fit,newdata=selected.genes)
Indeed, you're going to have to build the data frame of your selected genes.

Then see ?plot.survfit for additional plotting options. That'll help you to 
build a "plot(cox.surv,--other arguments--)" command to draw nice survival 
curves.

Hope this helps.

Jerome

On Tuesday 18 February 2003 15:04, Song, Guangchun wrote:
> Content-Length: 355
> Status: R
> X-Status: N
>
>
> I'm doing prediction of the survival cases using gene expression
> profiles(Affymetrix chips).  Can somebody tell me how to use the Cox PH
> model to select genes and make a prediction of survival?
>
>
> Thanks.
>
> Guangchun
>
> ______________________________________________
> [EMAIL PROTECTED] mailing list
> http://www.stat.math.ethz.ch/mailman/listinfo/r-help

-- 

Jerome Asselin (J�r�me)
Statistical Analyst
British Columbia Centre for Excellence in HIV/AIDS
St. Paul's Hospital
608 - 1081 Burrard Street
Vancouver, British Columbia
CANADA V6Z 1Y6

Email: [EMAIL PROTECTED]
Phone: 604 806-9112   Fax: 604 806-9044

______________________________________________
[EMAIL PROTECTED] mailing list
http://www.stat.math.ethz.ch/mailman/listinfo/r-help

Reply via email to