Prev 179719 / 398506 Next

problem with normalize.quantiles

Fri, May 8, 2009 6:47 AM

again, the bioconductor list is appropriate. You'll want to be sure to
update your packages using the instructions at

  http://bioconductor.org/docs/install/

and be sure to provide the output of

  sessionInfo()

to make any actual problems easier to track down.

Martin

George Chen <glchen at stanford.edu> writes:

Hi All,

I am using normalize.quantiles (package preprocessCore) to process a 8300 row x 96 col matrix, all numerical data.

When I limit the input matrix to 8300 x 67, the function works fine and my data is normalized.  But when I increase the input matrix size to 8300 x 68, the function processes but instead of giving normalized data, it gives me all the same number.  With 96 columns (the whole matrix), much the same happens.

Has anyone had any experience with this and what to do about it.  I would really like to not have to prune my dataset.

Examples:

Limited input matrix (col = 67)
          [,1]     [,2]     [,3]     [,4]     [,5]
 [1,]       NA 5977.433 5555.572 6826.831 4142.117
 [2,] 3012.837 6617.005 6036.862 7796.186 5453.295
 [3,]       NA 5977.433 7192.046 7582.068 6042.775
 [4,] 1780.039 1732.662 6373.366 7825.302 6775.492
 [5,]       NA 3308.601 2297.677 7294.935 1482.699
 [6,]       NA 3569.632 3509.979 3566.683       NA
 [7,]       NA 4492.529 6399.490 4925.750 1921.428
 [8,]       NA 2416.358 3958.061 4592.030 2829.436
 [9,] 7889.050 7928.198 7929.257 7930.121 7927.907
[10,] 7542.413 7893.827 7889.407 7896.078 7894.852

78 col input matrix
        [,1]     [,2]     [,3]     [,4]     [,5]
 [1,]       NA 7883.356 7883.356 7883.356 7883.356
 [2,] 7883.356 7883.356 7883.356 7883.356 7883.356
 [3,]       NA 7883.356 7883.356 7883.356 7883.356
 [4,] 7883.356 7883.356 7883.356 7883.356 7883.356
 [5,]       NA 7883.356 7883.356 7883.356 7883.356
 [6,]       NA 7883.356 7883.356 7883.356       NA
 [7,]       NA 7883.356 7883.356 7883.356 7883.356
 [8,]       NA 7883.356 7883.356 7883.356 7883.356
 [9,] 7883.356 7883.356 7883.356 7883.356 7883.356
[10,] 7883.356 7883.356 7883.356 7883.356 7883.356

whole input matrix (96 col)
         [,1]     [,2]     [,3]     [,4]     [,5]
 [1,]       NA 7747.377 7747.377 7747.377 7747.377
 [2,] 7747.377 7747.377 7747.377 7747.377 7747.377
 [3,]       NA 7747.377 7747.377 7747.377 7747.377
 [4,] 7747.377 7747.377 7747.377 7747.377 7747.377
 [5,]       NA 7747.377 7747.377 7747.377 7747.377
 [6,]       NA 7747.377 7747.377 7747.377       NA
 [7,]       NA 7747.377 7747.377 7747.377 7747.377
 [8,]       NA 7747.377 7747.377 7747.377 7747.377
 [9,] 7747.377 7747.377 7747.377 7747.377 7747.377
[10,] 7747.377 7747.377 7747.377 7747.377 7747.377

Thanks in advance,

George

______________________________________________
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.

Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M1 B861
Phone: (206) 667-2793

Thread (2 messages)

George Chen problem with normalize.quantiles May 7 Martin Morgan problem with normalize.quantiles May 8