fillPeaks ERROR

Topic: fillPeaks ERROR (Read 7903 times) previous topic - next topic

fillPeaks ERROR

December 30, 2011, 03:21:37 AM

Hi to all and good year!
I have had a problem running fillPeaks.
Controllo10 Error in checkSlotAssignment(object, name, value) : "scantime" is not a slot in class “NULL”

This are the settings I use:

Code: [Select]

xset <- xcmsSet(files, method="centWave", nSlaves=4, polarity="positive", ppm=5)
xset <- group(xset, bw=30, mzwid=0.025, minfrac=0.5, minsamp=10)
xset.retcor <- retcor(xset, method="obiwarp", profStep=0.1, distFunc= "cov", plottype="deviation")
xset.retcor <- group(xset.retcor, bw=10, mzwid=0.025, minfrac=0.5, minsamp=10)
xset.finale <- fillPeaks(xset.retcor)

My data set is made up of 60 negative control and 15 positive control. Is correct to use minsamp=10 that is 2/3 of the positive control? My data were aquired with Agilent LC/qTOF.

Best

Re: fillPeaks ERROR

Reply #1 – December 30, 2011, 01:10:45 PM

please read before posting!

We need at least
sessionInfo()
and
traceback()

Re: fillPeaks ERROR

Reply #2 – January 03, 2012, 02:51:18 AM

Quote

please read before posting!

:oops:

The error is:

xsetF <- fillPeaks(xsetR)
Controllo10 Error in checkSlotAssignment(object, name, value) : "scantime" is not a slot in class “NULL”

Furthemore:

> sessionInfo()
R version 2.14.1 (2011-12-22)
Platform: x86_64-pc-linux-gnu (64-bit)

locale:
[1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C
[3] LC_TIME=en_US.utf8 LC_COLLATE=en_US.utf8
[5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8
[7] LC_PAPER=C LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C

attached base packages:
[1] grid stats graphics grDevices utils datasets methods
[8] base

other attached packages:
[1] ROCR_1.0-4 gplots_2.10.1 KernSmooth_2.23-7 caTools_1.12
[5] bitops_1.0-4.1 gdata_2.8.2 gtools_2.6.2 boot_1.3-3
[9] chemometrics_1.3.7 som_0.3-5 robustbase_0.7-8 rpart_3.1-50
[13] pls_2.2-0 pcaPP_1.9-44 mvtnorm_0.9-9991 nnet_7.3-1
[17] mclust_3.4.10 MASS_7.3-16 lars_0.9-8 gclus_1.3
[21] cluster_1.14.1 e1071_1.6 class_7.3-3 CAMERA_1.11.1
[25] snow_0.3-7 xcms_1.30.3

loaded via a namespace (and not attached):
[1] graph_1.32.0 Hmisc_3.9-0 igraph_0.5.5-3 lattice_0.20-0 RBGL_1.30.1
[6] tools_2.14.1

and:
> traceback()
11: stop(gettextf(""%s" is not a slot in class %s", name, dQuote(cl)),
domain = NA)
10: checkSlotAssignment(object, name, value)
9: `slot<-`(object, name, TRUE, value)
8: `@<-`(`*tmp*`, scantime, value = c(364.497009277344, 365.298553466797,
366.100128173828, 366.901794433594, 367.703460693359, 368.505187988281, .... many more.....
2700.44604492188, 2701.31811523438, 2702.17993164062))
7: .local(object, ...)
6: fillPeaks.chrom(object, ...)
5: fillPeaks.chrom(object, ...)
4: do.call(method, alist(object, ...))
3: .local(object, ...)
2: fillPeaks(xsetR)
1: fillPeaks(xsetR)
>

Best

P.S. Happy new year!! :mrgreen:

Re: fillPeaks ERROR

Reply #3 – January 03, 2012, 05:15:55 PM

Very strange. For some reason this assignment

Code: [Select]

lcraw@scantime <- rtcor[[i]]

in fillPeaks.chrom() seems to fail in that case.

Does it work on XCMS Online ?

By the way, there is no need to pre-group if you use OBIWarp.

Re: fillPeaks ERROR

Reply #4 – January 04, 2012, 02:22:56 AM

Quote

Does it work on XCMS Online ?

No, I work on my worrksation.

Quote

there is no need to pre-group if you use OBIWarp

Thanks for the suggestion.

How can I go on with my analysis?

Best

Re: fillPeaks ERROR

Reply #5 – January 04, 2012, 11:06:56 AM

What I meant was have you tried analysing the same dataset on XCMS Online,
to see if it works there ?

Re: fillPeaks ERROR

Reply #6 – January 04, 2012, 01:26:54 PM

Since we don't know what effects the brand-new R2.14.1 might have on XCMS,
you might want to try using a well-tested version like R2.12.1.

Re: fillPeaks ERROR

Reply #7 – January 04, 2012, 03:10:39 PM

How can I downgrade R 2.14.1 to R 2.14.1??

What could happens to all the library installed?? Should I re-install all??

Re: fillPeaks ERROR

Reply #8 – January 11, 2012, 11:05:10 AM

Quote from: "Ralf"

What I meant was have you tried analysing the same dataset on XCMS Online,
to see if it works there ?

Hi I tried to use xcms online and it works. So the problem is in my workstation. Have you ever tried to use xcms with R 2.14?? Uninstall and re-install the downgraded version of R is the very last option. :cry:

Do you have any further suggestion??

Best

EDIT
I also tried on my home computer and all works fine.
The session info on my home pc is:

> traceback()
9: logical(nrow(m))
8: rectUnique(groupmat[, c("mzmin", "mzmax", "rtmin", "rtmax"),
drop = FALSE], uorder)
7: .local(object, ...)
6: group.density(object, ...)
5: group.density(object, ...)
4: do.call(method, alist(object, ...))
3: .local(object, ...)
2: group(xset, bw = 30, mzwid = 0.025, minfrac = 0.5, minsamp = 10)
1: group(xset, bw = 30, mzwid = 0.025, minfrac = 0.5, minsamp = 10)

> sessionInfo( )
R version 2.14.1 (2011-12-22)
Platform: x86_64-pc-linux-gnu (64-bit)

locale:
[1] LC_CTYPE=it_IT.utf8 LC_NUMERIC=C
[3] LC_TIME=it_IT.utf8 LC_COLLATE=it_IT.utf8
[5] LC_MONETARY=it_IT.utf8 LC_MESSAGES=it_IT.utf8
[7] LC_PAPER=C LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=it_IT.utf8 LC_IDENTIFICATION=C

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] boot_1.3-3 chemometrics_1.3.7 som_0.3-5 robustbase_0.7-8
[5] rpart_3.1-50 pls_2.2-0 pcaPP_1.9-44 mvtnorm_0.9-9991
[9] nnet_7.3-1 mclust_3.4.10 MASS_7.3-16 lars_0.9-8
[13] gclus_1.3 cluster_1.14.1 e1071_1.6 class_7.3-3
[17] CAMERA_1.10.0 xcms_1.30.0

loaded via a namespace (and not attached):
[1] graph_1.32.0 grid_2.14.1 Hmisc_3.9-0 igraph_0.5.5-3 lattice_0.20-0
[6] RBGL_1.30.1 tools_2.14.1
>
I see the main difference is the xcms version that in the home pc is 1.30.0.

Where can I download these old version to try it on my work pc??

Re: fillPeaks ERROR

Reply #9 – January 12, 2012, 06:34:36 AM

Quote from: "Ricca"

Where can I download these old version to try it on my work pc??

If you don't have time to wait for the bugfix or want to reinstall,
you could download an older version (1.26.1) from the BioC 2.8 release (http://bioconductor.org/packages/2.8/bi ... /xcms.html).

If you need the exact 1.30.0 and you have a little experience with R and svn, you could do a direct checkout from the Bioconductor svn tree (see http://www.bioconductor.org/developers/source-control/).

You need Revision 59920, which can be checked out in a linux terminal with:
svn checkout https://hedgehog.fhcrc.org/bioconductor ... xcms@59920

Afterwards a "R CMD install xcms" installs the package, if you have all dependencies.

Carsten