[BioC] contrasts.fit Error

Fri Sep 24 18:14:25 CEST 2004

You design matrix singular.  If you run lmFit() it will warn you that coefficient
I is non-estimable.

Gordon

> I updated my version and still get the same error. Here is the model
> matrix if that helps:
> cln.mm
>      A B C D E F H I J
> 1   0 0 0 1 0 1 0 0 0
> 2   0 0 0 1 1 0 0 0 0
> 3   1 0 0 0 0 1 0 0 0
> 4   0 0 1 0 0 0 0 1 0
> 5   0 0 1 0 0 0 0 1 0
> 6   0 0 0 1 0 1 0 0 0
> 7   1 0 0 0 1 0 0 0 0
> 8   0 0 0 1 1 0 0 0 0
> 9   1 0 0 0 0 0 1 0 0
> 10  0 0 1 0 0 0 0 1 0
> 11  1 0 0 0 1 0 0 0 0
> 12  0 1 0 0 0 0 0 1 0
> 13  0 1 0 0 0 1 0 0 0
> 14  1 0 0 0 1 0 0 0 0
> 15  1 0 0 0 0 0 0 1 0
> 16  0 0 1 0 0 0 0 1 0
> 17  0 0 0 1 1 0 0 0 0
> 18  1 0 0 0 0 0 1 0 0
> 19  1 0 0 0 0 0 1 0 0
> 20  0 0 0 1 1 0 0 0 0
> 21  1 0 0 0 1 0 0 0 0
> 22  0 0 0 1 1 0 0 0 0
> 23  0 0 0 1 0 0 1 0 0
> 24  0 0 0 1 0 0 1 0 0
> 25  0 0 1 0 0 0 0 1 0
> 26  0 0 0 1 0 0 1 0 0
> 27  1 0 0 0 1 0 0 0 0
> 28  1 0 0 0 0 0 1 0 0
> 29  0 0 1 0 0 0 0 1 0
> 30  1 0 0 0 0 0 1 0 0
> 110 0 0 1 0 1 0 0 0 1
> 210 0 0 1 0 1 0 0 0 1
> 31  1 0 0 0 0 1 0 0 1
> 41  1 0 0 0 0 0 0 1 1
> 51  1 0 0 0 0 0 0 1 1
> 61  0 1 0 0 0 1 0 0 1
> 71  0 0 0 1 0 0 0 1 1
> 81  1 0 0 0 1 0 0 0 1
> 91  0 1 0 0 0 0 1 0 1
> 101 1 0 0 0 0 0 0 1 1
> 111 1 0 0 0 0 1 0 0 1
> 121 0 1 0 0 1 0 0 0 1
> 131 0 1 0 0 0 1 0 0 1
> 141 1 0 0 0 1 0 0 0 1
> 151 0 1 0 0 1 0 0 0 1
> 161 0 1 0 0 0 0 1 0 1
> 171 1 0 0 0 1 0 0 0 1
> 181 0 1 0 0 1 0 0 0 1
> 191 0 1 0 0 1 0 0 0 1
> 201 1 0 0 0 0 0 0 1 1
> 211 0 1 0 0 0 1 0 0 1
> 221 0 1 0 0 0 0 0 1 1
> 231 1 0 0 0 0 0 0 1 1
>
> cln.cm <- makeContrasts(A-B, A-C, B-C, levels=cln.mm)
> cln.fit2 <- contrasts.fit(cln.fit, cln.cm)
>
> I'm now running limma version 1.7.7
> Thanks for any help you can give.
> Matt
>
> On Sep 24, 2004, at 11:22 AM, Gordon K Smyth wrote:
>
>> We have seen this error message previously when the design matrix was
>> singular and people
>> attempted to take contrasts of non-estimable coefficients.  There was
>> a fix introduced in limma
>> 1.7.5 to warn about non-singular design matrices and to catch this
>> problem in contrasts.fit().
>>
>> Try updating limma and see if the above turns out to be the problem.
>> It is easiest to update
>> limma directly from CRAN, just typing
>>
>>> install.packages("limma")
>>
>> should do it.
>>
>> Gordon
>>
>>> When I try to run the following line of code:
>>> cln.fit2 <- contrasts.fit(cln.fit, cln.cm)
>>> where cln.fit comes from:
>>> cln.fit <- lmFit(colon$M, cln.mm)
>>> where colon is a limma object and cln.mm is the model matrix.
>>> also cln.cm is the contrast matrix from:
>>> cln.cm <- makeContrasts(MSLiver-MSLung, MSLiver-MSLiverLung,
>>> MSLung-MSLiverLung, levels=cln.mm)
>>>
>>> I get the error:
>>> Error in o %*% RUC^2 : non-conformable arguments
>>>
>>> Can anyone tell me what I'm doing wrong?
>>>
>>> I'm running:
>>> Package: limma
>>> Version: 1.7.0
>>> Date: 2004/06/27
>>> Title: Linear Models for Microarray Data
>>> Author: Gordon Smyth <smyth at wehi.edu.au>, Matt Ritchie
>>> <mritchie at wehi.edu.au>, James Wettenhall <wettenhall at wehi.edu.au>,
>>> Natalie Thorne
>>>                  <thorne at wehi.edu.au>
>>> Maintainer: Gordon Smyth <smyth at wehi.edu.au>
>>> Depends: R (>= 1.7.1), MASS, splines, statmod (>= 1.0.6), sma
>>> Description: Data analysis, linear models and differential expression
>>> for microarray data.
>>> License: LGPL
>>> URL: http://bioinf.wehi.edu.au/limma/
>>> Packaged: Tue Jun 29 21:38:42 2004; madman
>>> Built: R 2.0.0; ; 2004-06-30 13:15:39; windows
>>>
>>> platform powerpc-apple-darwin6.8
>>> arch     powerpc
>>> os       darwin6.8
>>> system   powerpc, darwin6.8
>>> status
>>> major    1
>>> minor    9.1
>>> year     2004
>>> month    06
>>> day      21
>>> language R
>>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
>