Question

gls from nlme package not compatible with methods package

0

Entering edit mode

James Wettenhall ▴ 1000

@james-wettenhall-153

Last seen 10.2 years ago

Dear Bioconductor, I am trying to develop a Bioconductor-style package for fitting linear models to microarray data, to be used with marrayNorm, marrayInput etc. I want to use the function gls (generalized least squares) from the nlme package. But it doesn't seem to be compatible with the methods package. Any suggestions? Regards, James > y <- rnorm(20) > x <- 1:20 > library(nlme) Loading required package: nls > out <- gls(y~x) > summary(out) Generalized least squares fit by REML Model: y ~ x Data: NULL AIC BIC logLik 63.18472 65.85584 -28.59236 Coefficients: Value Std.Error t-value p-value (Intercept) -0.4416286 0.4227542 -1.044646 0.3100 x -0.0004764 0.0352908 -0.013499 0.9894 Correlation: (Intr) x -0.877 Standardized residuals: Min Q1 Med Q3 Max -1.85167510 -0.68224481 0.03176228 0.71367684 1.79529312 Residual standard error: 0.9100658 Degrees of freedom: 20 total; 18 residual > out <- gls(y~x) > library(methods) > out <- gls(y~x) Error in getClass(Class) : "glsStruct" is not a defined class ---------------------------------------------------------------------- ---- James Wettenhall Tel: (+61 3) 9345 2629 Division of Genetics and Bioinformatics Fax: (+61 3) 9347 0852 The Walter & Eliza Hall Institute E-mail: wettenhall@wehi.edu.au of Medical Research, Mobile: (+61 / 0 ) 438 527 921 1G Royal Parade, Parkville, Vic 3050, Australia http://www.wehi.edu.au ---------------------------------------------------------------------- ----

Microarray Genetics Microarray Genetics • 1.6k views

ADD COMMENT • link updated 21.9 years ago by Douglas Bates ▴ 180 • written 21.9 years ago by James Wettenhall ▴ 1000

score 0 · Answer 1 · 2003-01-13

0

Entering edit mode

Douglas Bates ▴ 180

@douglas-bates-5

Last seen 10.2 years ago

James Wettenhall <wettenhall@wehi.edu.au> writes: > Dear Bioconductor, > > I am trying to develop a Bioconductor-style package for fitting > linear models to microarray data, to be used with marrayNorm, > marrayInput etc. > > I want to use the function gls (generalized least squares) from > the nlme package. But it doesn't seem to be compatible with the > methods package. Any suggestions? ... > > out <- gls(y~x) > > library(methods) > > out <- gls(y~x) > Error in getClass(Class) : "glsStruct" is not a defined class What version of R and what version of the nlme package? Please send the result of R.version and the first few lines of library(help=nlme). It should look like > R.version _ platform i386-pc-linux-gnu arch i386 os linux-gnu system i386, linux-gnu status major 1 minor 6.2 year 2003 month 01 day 10 language R > library(help=nlme) nlme Linear and nonlinear mixed effects models Description: Package: nlme Version: 3.1-36 Date: 2002-12-28 Priority: recommended Title: Linear and nonlinear mixed effects models Also, it would help to send the results of running traceback() immediately after you get the error message.

ADD COMMENT • link 21.9 years ago Douglas Bates ▴ 180

0

Entering edit mode

Douglas, I just tried R 1.6.1 in Windows with methods 1.6.1 and nlme 3.1-33 and there was no problem :) But when running on Linux with R 1.5.0 and methods 1.5.0 and nlme 3.1-23 I got the error described previously. Maybe I'll have to tell my sysadmin to upgrade R ! Regards, James > R.version _ platform i686-pc-linux-gnu arch i686 os linux-gnu system i686, linux-gnu status major 1 minor 5.0 year 2002 month 04 day 29 language R > library(help=nlme) nlme Linear and nonlinear mixed effects models Description: Package: nlme Version: 3.1-23 Date: 2001/01/28 > out<-gls(y~x) Error in getClass(Class) : "glsStruct" is not a defined class > traceback() 4: stop(paste("\"", Class, "\" is not a defined class", sep = "")) 3: getClass(Class) 2: initialize(glsSt, dataMod, glsEstControl) 1: gls(y ~ x) On 13 Jan 2003, Douglas Bates wrote: > James Wettenhall <wettenhall@wehi.edu.au> writes: > > > Dear Bioconductor, > > > > I am trying to develop a Bioconductor-style package for fitting > > linear models to microarray data, to be used with marrayNorm, > > marrayInput etc. > > > > I want to use the function gls (generalized least squares) from > > the nlme package. But it doesn't seem to be compatible with the > > methods package. Any suggestions? > > ... > > > > out <- gls(y~x) > > > library(methods) > > > out <- gls(y~x) > > Error in getClass(Class) : "glsStruct" is not a defined class > > What version of R and what version of the nlme package? > > Please send the result of R.version and the first few lines of > library(help=nlme). It should look like > > > R.version > _ > platform i386-pc-linux-gnu > arch i386 > os linux-gnu > system i386, linux-gnu > status > major 1 > minor 6.2 > year 2003 > month 01 > day 10 > language R > > library(help=nlme) > nlme Linear and nonlinear mixed effects models > > Description: > > Package: nlme > Version: 3.1-36 > Date: 2002-12-28 > Priority: recommended > Title: Linear and nonlinear mixed effects models > > Also, it would help to send the results of running traceback() > immediately after you get the error message. > > -- ---------------------------------------------------------------------- ---- James Wettenhall Tel: (+61 3) 9345 2629 Division of Genetics and Bioinformatics Fax: (+61 3) 9347 0852 The Walter & Eliza Hall Institute E-mail: wettenhall@wehi.edu.au of Medical Research, Mobile: (+61 / 0 ) 438 527 921 1G Royal Parade, Parkville, Vic 3050, Australia http://www.wehi.edu.au ---------------------------------------------------------------------- ----

ADD REPLY • link 21.9 years ago James Wettenhall ▴ 1000