I removed those plmData directories, but I still had the same error.
I also tried running it on a computer with an older installation of
aroma.affymetrix (since I have run this previously on this same data
and it has worked), but with the same error. I do notice that it
shows that all the fragment lengths are NA:
Finished reading CEL file.
20100823 11:06:13| Thetas:
num [1:233566, 1] 971 811 1509 1323 486 ...
20100823 11:06:13| (Allele-specific) thetas:
num [1:233566, 1] 971 811 1509 1323 486 ...
20100823 11:06:13| Total thetas:
num [1:233566] 971 811 1509 1323 486 ...
20100823 11:06:13| Summary of total signals (on the intensity
scale):
Min. 1st Qu. Median Mean 3rd Qu. Max.
389 3780 5670 6700 8510 62600
20100823 11:06:14| Signals:
num [1:233566] 9.92 9.66 10.56 10.37 8.93 ...
20100823 11:06:14| Summary of signals (on the log2 scale):
Min. 1st Qu. Median Mean 3rd Qu. Max.
8.6 11.9 12.5 12.5 13.1 15.9
20100823 11:06:14| Fragment lengths:
int [1:233566, 1] NA NA NA NA NA NA NA NA NA NA ...
20100823 11:06:14| Summary of fragment lengths:
V1
Min. : NA
1st Qu.: NA
Median : NA
Mean : NaN
3rd Qu.: NA
Max. : NA
NA's :233566
20100823 11:06:14| Fitting target prediction function to each enzyme
exclusively...
20100823 11:06:14| Distribution of log2 signals that are finite:
Mode TRUE NA's
logical 233566 0
20100823 11:06:14| Distribution of units with known fragment
lengths:
V1
Mode :logical
FALSE:233566
NA's :0
20100823 11:06:14| Enzyme #1 of 1...
20100823 11:06:14| Distribution of units with known fragment
lengths and finite signals:
Mode FALSE NA's
logical 233566 0
20100823 11:06:14| Distribution of units with known fragment
lengths and finite signals that are exclusively on this enzyme:
Mode FALSE NA's
logical 233566 0
Error in list(`lapply(chipTypes, function(x) doCRMA("gsk", chipType =
x, verbose = verbose` = <environment>, :
Gene
On Aug 19, 8:39 pm, Henrik Bengtsson <henrik.bengts...@gmail.com>
wrote:
> Hi Gene,
>
> did you add additional arrays to your rawData/ directory at anytime
> after having ran the pipeline ones? If so, that would explain the
> problem. The solution is then to delete the two plmData/
> subdirectories:
>
> plmData/gsk,ACC,-XY,BPN,-XY,RMA,A+B/Mapping250K_Sty/
> plmData/gsk,ACC,-XY,BPN,-XY,RMA,A+B,FLN,-XY/Mapping250K_Sty/
>
> which we give you new probe-level model estimates. The thing is that
> RmaCnPlm is a multi-array method, meaning that any estimates has to be
> redone if you add new arrays. Since the fragment-length normalized
> data depend on this too, you have to reestimate those as well. The
> easiest way is to delete the corresponding plmData/ directories as
> above.
>
> Hope this helps
>
> Henrik
>
> PS. Technically, you've detected a bug in RmaPlm, because it fails to
> detect that new arrays have been added. Instead, it quietly pass
> empty allocated chip-effect estimates to FragmentLengthNormalization
> who has to deal with it ...and luckily detected it.
>
>
>
--
When reporting problems on aroma.affymetrix, make sure 1) to run the latest
version of the package, 2) to report the output of sessionInfo() and
traceback(), and 3) to post a complete code example.
You received this message because you are subscribed to the Google Groups
"aroma.affymetrix" group with website http://www.aroma-project.org/.
To post to this group, send email to aroma-affymetrix@googlegroups.com
To unsubscribe and other options, go to http://www.aroma-project.org/forum/
