R help - Aug 2004 - A question about external time-dependent covariates in co x model

Dear Rui,
>From my understanding of time-dependent covariates (not an expert but have
been working on a similar problem), it would appear that the coding of the
status column is not correct. Unless you have observed an event at each
interval you should only have status=1 for the last interval. In your
example I see 3 in total. Also, I think that if "end" is proportional
to
your "covariate" you are incorporating a redundant time effect into
the
model. The time effect is in the baseline hazard.

Alex
-----Original Message-----
From: Rui Song [mailto:rsong at stat.wisc.edu] 
Sent: August 19, 2004 12:21 AM
To: r-help at stat.math.ethz.ch
Subject: [R] A question about external time-dependent covariates in cox
model


Dear Sir or Madam:
I am a graduate student in UW-Madison statistics department. I have a
question about fitting a cox model with external time-dependent
covariates.

Say the original data is in the following format:
Obs Eventtime  Status  Cov(time=5)  Cov(time=8)  Cov(time=10)	Cov(time=12)
1	5	1		2
2	8	0(censored)	2	4
3	10	1		2	4		6
4	12	1		2	4		6		8
....

Notice that the time-dependent covariates are identical at the same
time points for all obs since they are external to the failure process.
process.

Then I organized the data as the following:
obs	start	end	eventtime	status	cov
1	0	5	5		1	2
2	0	5	8		0	2
2	5	8	8		0	4
3	0	5	10		1	2
3	5	8	10		1	4
3	8	10	10		1	6
4	0	5	12		1	2
4	5	8	12		1	4
4	8	10	12		1	6
4	10	12	12		1	8

And fit the model using:

fit<-coxph(Surv(start, end, status)~cov);

When I fit the model to my data set (Which has 89 observations and 81
distinct time points, sort of large.), I always got a message that
"Process R segmentation fault (core dumped)". Would you let me know if
it
is due to the matrix sigularity in the computation of the partial
likelihood or something else? And how should I fit a cox model with
external time-dependent covariates?

Thanks a lot for your time and help!

Sincerely,
Rui Song

______________________________________________
R-help at stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

Dear Alex,

Thanks a lot for your help! The "end" is not proportional to the
covariate in my data set. I recoded the status as you suggested, but the
problem still exists. I tried to fit the model with partial data, it has
no problem when I use the first 25 rows (from the data in the form
"start", "end", "status", "cov"), with
very huge standard error. And the
problem exists if I cooperated more data. It is kind of weird...

Rui

On Thu, 19 Aug 2004, Hanke, Alex wrote:
> Dear Rui,
> >From my understanding of time-dependent covariates (not an expert but
have
> been working on a similar problem), it would appear that the coding of the
> status column is not correct. Unless you have observed an event at each
> interval you should only have status=1 for the last interval. In your
> example I see 3 in total. Also, I think that if "end" is
proportional to
> your "covariate" you are incorporating a redundant time effect
into the
> model. The time effect is in the baseline hazard.
>
> Alex
> -----Original Message-----
> From: Rui Song [mailto:rsong at stat.wisc.edu]
> Sent: August 19, 2004 12:21 AM
> To: r-help at stat.math.ethz.ch
> Subject: [R] A question about external time-dependent covariates in cox
> model
>
>
> Dear Sir or Madam:
> I am a graduate student in UW-Madison statistics department. I have a
> question about fitting a cox model with external time-dependent
> covariates.
>
> Say the original data is in the following format:
> Obs Eventtime  Status  Cov(time=5)  Cov(time=8)  Cov(time=10)	Cov(time=12)
> 1	5	1		2
> 2	8	0(censored)	2	4
> 3	10	1		2	4		6
> 4	12	1		2	4		6		8
> ....
>
> Notice that the time-dependent covariates are identical at the same
> time points for all obs since they are external to the failure process.
> process.
>
> Then I organized the data as the following:
> obs	start	end	eventtime	status	cov
> 1	0	5	5		1	2
> 2	0	5	8		0	2
> 2	5	8	8		0	4
> 3	0	5	10		1	2
> 3	5	8	10		1	4
> 3	8	10	10		1	6
> 4	0	5	12		1	2
> 4	5	8	12		1	4
> 4	8	10	12		1	6
> 4	10	12	12		1	8
>
> And fit the model using:
>
> fit<-coxph(Surv(start, end, status)~cov);
>
> When I fit the model to my data set (Which has 89 observations and 81
> distinct time points, sort of large.), I always got a message that
> "Process R segmentation fault (core dumped)". Would you let me
know if it
> is due to the matrix sigularity in the computation of the partial
> likelihood or something else? And how should I fit a cox model with
> external time-dependent covariates?
>
> Thanks a lot for your time and help!
>
> Sincerely,
> Rui Song
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
>

Dear group, 
Apologies for asking the most chomped FAQ.

I have a file with list of gene names(genes.txt):
EGF
EGFR
PTPN6
TIEG2
MAPK1


I have another object in R, I do not know the data
type for that object that looks like this
("lidnames"):
"RABGGTA"   "MAPK3"     "TIE" "CYP2C19" 

> lidnames[1:10]
 100_g_at   1000_at   1001_at 1002_f_at 1003_s_at    
"RABGGTA"   "MAPK3"     "TIE" "CYP2C19" 
"BLR1"


I want to pick list of genes from lidnames object that
are in genes.txt.  I am using %in% function.
>mygenes<-read.table("genes.txt")
> mygenes[mygenes %in% lidnames]
NULL data frame with 164 rows
> 
I am unable to pullout genes from lidnames object.

Is it because that lidnames is as a list type and
mygenes object is as a vector/matrix type.

How can I convert mygenes to list type where I can
have the elements:

"EGF","EGFR","PTPN6","TIEG2","MAPK1".

I think in this way I can pull out the names from
lidnames object or any other complex matrix. 

Please help me. 

Thank you
SP

S Peri wrote:
> Dear group, 
> Apologies for asking the most chomped FAQ.
> 
> I have a file with list of gene names(genes.txt):
> EGF
> EGFR
> PTPN6
> TIEG2
> MAPK1
> 
> 
> I have another object in R, I do not know the data
> type for that object that looks like this
> ("lidnames"):
> "RABGGTA"   "MAPK3"     "TIE"
"CYP2C19"
> 
> 
> 
>>lidnames[1:10]
> 
>  100_g_at   1000_at   1001_at 1002_f_at 1003_s_at    
> "RABGGTA"   "MAPK3"     "TIE"
"CYP2C19"    "BLR1"
> 
> 
> I want to pick list of genes from lidnames object that
> are in genes.txt.  I am using %in% function.
> 
> 
>>mygenes<-read.table("genes.txt")
> 
> 
>>mygenes[mygenes %in% lidnames]
> 
> NULL data frame with 164 rows
> 
> 
> I am unable to pullout genes from lidnames object.
> 
> Is it because that lidnames is as a list type and
> mygenes object is as a vector/matrix type.
Well, mygenes is a data frame, and lidnames a vector.
E.g., try to read in mygenes as a vector directly:
  mygenes <- scan("genes.txt", what = "character")

Uwe Ligges
> How can I convert mygenes to list type where I can
> have the elements:
> 
>
"EGF","EGFR","PTPN6","TIEG2","MAPK1".
> 
> I think in this way I can pull out the names from
> lidnames object or any other complex matrix. 
> 
> Please help me. 
> 
> Thank you
> SP
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

On Thu, Aug 19, 2004 at 09:36:22AM -0300, Hanke, Alex
wrote:
> Dear Rui,
> >From my understanding of time-dependent covariates (not an expert but
have
> been working on a similar problem), it would appear that the coding of the
> status column is not correct. Unless you have observed an event at each
> interval you should only have status=1 for the last interval. In your
> example I see 3 in total. Also, I think that if "end" is
proportional to
> your "covariate" you are incorporating a redundant time effect
into the
> model. The time effect is in the baseline hazard.
Right, the 'splitting' was made incorrectly, but 'coxph'
shouldn't
segfault anyway. The error seems to be (caught) in 'coxph_wtest.c',
line 29, which may be of interest to the R maintainer of 'survival', 
Thomas L.

G??ran
> 
> Alex
> -----Original Message-----
> From: Rui Song [mailto:rsong at stat.wisc.edu] 
> Sent: August 19, 2004 12:21 AM
> To: r-help at stat.math.ethz.ch
> Subject: [R] A question about external time-dependent covariates in cox
> model
> 
> 
> Dear Sir or Madam:
> I am a graduate student in UW-Madison statistics department. I have a
> question about fitting a cox model with external time-dependent
> covariates.
> 
> Say the original data is in the following format:
> Obs Eventtime  Status  Cov(time=5)  Cov(time=8)  Cov(time=10)	Cov(time=12)
> 1	5	1		2
> 2	8	0(censored)	2	4
> 3	10	1		2	4		6
> 4	12	1		2	4		6		8
> ....
> 
> Notice that the time-dependent covariates are identical at the same
> time points for all obs since they are external to the failure process.
> process.
> 
> Then I organized the data as the following:
> obs	start	end	eventtime	status	cov
> 1	0	5	5		1	2
> 2	0	5	8		0	2
> 2	5	8	8		0	4
> 3	0	5	10		1	2
> 3	5	8	10		1	4
> 3	8	10	10		1	6
> 4	0	5	12		1	2
> 4	5	8	12		1	4
> 4	8	10	12		1	6
> 4	10	12	12		1	8
> 
> And fit the model using:
> 
> fit<-coxph(Surv(start, end, status)~cov);
> 
> When I fit the model to my data set (Which has 89 observations and 81
> distinct time points, sort of large.), I always got a message that
> "Process R segmentation fault (core dumped)". Would you let me
know if it
> is due to the matrix sigularity in the computation of the partial
> likelihood or something else? And how should I fit a cox model with
> external time-dependent covariates?
> 
> Thanks a lot for your time and help!
> 
> Sincerely,
> Rui Song
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
-- 
 G??ran Brostr??m                    tel: +46 90 786 5223
 Department of Statistics          fax: +46 90 786 6614
 Ume?? University                   http://www.stat.umu.se/egna/gb/
 SE-90187 Ume??, Sweden             e-mail: gb at stat.umu.se

Peri, 

Please stop posting to both R-help and BioConductor simultaneously. 

Please read the posting guide and decide which list is more appropriate.
http://www.R-project.org/posting-guide.html





On Thu, 2004-08-19 at 17:20, S Peri wrote:
> Dear group, 
> Apologies for asking the most chomped FAQ.
> 
> I have a file with list of gene names(genes.txt):
> EGF
> EGFR
> PTPN6
> TIEG2
> MAPK1
> 
> 
> I have another object in R, I do not know the data
> type for that object that looks like this
> ("lidnames"):
> "RABGGTA"   "MAPK3"     "TIE"
"CYP2C19"
> 
> 
> > lidnames[1:10]
>  100_g_at   1000_at   1001_at 1002_f_at 1003_s_at    
> "RABGGTA"   "MAPK3"     "TIE"
"CYP2C19"    "BLR1"
> 
> 
> I want to pick list of genes from lidnames object that
> are in genes.txt.  I am using %in% function.
> 
> >mygenes<-read.table("genes.txt")
> 
> > mygenes[mygenes %in% lidnames]
> NULL data frame with 164 rows
> > 
> 
> I am unable to pullout genes from lidnames object.
> 
> Is it because that lidnames is as a list type and
> mygenes object is as a vector/matrix type.
> 
> How can I convert mygenes to list type where I can
> have the elements:
> 
>
"EGF","EGFR","PTPN6","TIEG2","MAPK1".
> 
> I think in this way I can pull out the names from
> lidnames object or any other complex matrix. 
> 
> Please help me. 
> 
> Thank you
> SP
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>

Dear S,

when posting something to R-help or bioconductor,
please do *NOT* reply to an existing message and add a new
subject!!!  {it breaks threads for most e-mail system that use
	     threads, including the threaded archives, see, e.g.
   https://stat.ethz.ch/pipermail/bioconductor/2004-August/thread.html
   (and look for "S Peri").

This is clearly said in the posting guide, see below
>>>>> "S" == S Peri <biocperi at yahoo.com>
>>>>>     on Thu, 19 Aug 2004 09:20:41 -0700 (PDT) writes:
    S> Dear group, 
    
  <..........>

    S> ______________________________________________
    S> R-help at stat.math.ethz.ch mailing list
    S> https://stat.ethz.ch/mailman/listinfo/r-help
    S> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
      
PLEASE do read it!

Martin Maechler

Hi Kevin,
 {and R-help readers:  Kevin's (privatae) question is a quite
  relevant one to this sidetrack topic of "mailing list netiquette"}
>>>>> "Kevin" == Kevin Wang <Kevin.Wang at
maths.anu.edu.au>
>>>>>     on Fri, 20 Aug 2004 22:37:39 +1000 (EST) writes:
    Kevin> Just out of interest, what about those posts with (no
    Kevin> subject) as the subject?  Sometimes I would change
    Kevin> the subject so it fits the body...is it recommended?

yes, I'd recommend it --- IFF --- your e-mail system is good
enough to set / augment the (usually invisible) headers
"References:" and "In-Reply-To:".

AFAIK, by official standards, these are used for threading
(if available) rather than the subjects.

Martin

    Kevin> On Fri, 20 Aug 2004, Martin Maechler wrote:

    >> Dear S,
    >> 
    >> when posting something to R-help or bioconductor,
    >> please do *NOT* reply to an existing message and add a new
    >> subject!!!  {it breaks threads for most e-mail system that use
    >> threads, including the threaded archives, see, e.g.
    >> https://stat.ethz.ch/pipermail/bioconductor/2004-August/thread.html