[BioC] Newbie 'rflowcyt' question; problem loading facscan256.fcs

Ulrik Stervbo ulriks at ruc.dk
Thu Jun 1 06:58:24 CEST 2006 

Hi Seth, Hi Nolwenn,

I installed the latest rflowcyt, and get a different error:
> fcs.loc <- system.file("fcs", package = "rfcdmin")
> file.location <- paste(fcs.loc, "facscan256.fcs", sep = "/")
> FC.FCSRobj <- read.FCS(file.location, UseS3 = TRUE, MY.DEBUG = TRUE,fcs.type = NULL)
		Start/Stop:$version
		[1] "FCS2.0"
		
		$TextStart
		[1] 256
		
		$TextEnd
		[1] 1595
		
		$TextLength
		[1] 1340
		
		$DataStart
		[1] 1792
		
		$DataEnd
		[1] 51791
		
		$DataLength
		[1] 50000
		
		$AnalysisStart
		[1] 0
		
		$AnalysisEnd
		[1] 0
		
		Error in 1:last.pos.delim : result would be too long a vector
		In addition: Warning messages:
		1: no non-missing arguments to max; returning -Inf
		2: Metadata Length Discrepancy: TextLength = 1340 but valid metadata
length = -Inf in: read.FCS(file.location, UseS3 = TRUE, MY.DEBUG =
TRUE, fcs.type = NULL)
		3: The metadata will be truncated up to the last delimiter. in:
read.FCS(file.location, UseS3 = TRUE, MY.DEBUG = TRUE, fcs.type =
NULL)

The result form library(help=rflowcyt):
Information on package 'rflowcyt'

Description:

Package:       rflowcyt
Version:       1.5.2
Date:          2006-05-27
Title:         Statistical tools and data structures for analytic flow
cytometry
Author:        A.J. Rossini, J.Y. Wan, N. Le Meur and Zoe Moodie
Maintainer:    N. Le Meur <nlemeur at fhcrc.org>
	Depends:       R (>= 2.3.0), survival, xtable, stats, KernSmooth,
	MASS,grid, splancs, methods, hexbin, rfcdmin (>= 0.9.0),
	fields, lattice,locfit
	Imports:       prada,graphics
	Suggests:      xgobi
	Description:   Provides basic S4 data structures and routines for
	analysing flow cytometry data.
	License:       GNU General Public License 2.0.
	URL:           http://www.analytics.washington.edu/downloads/rflowcyt/
	biocViews:     Statistics, Infrastructure, CellBasedAssays
	Packaged:      Sun May 28 05:37:33 2006; biocbuild
Built:         R 2.3.0; ; 2006-05-31 17:17:31; unix

And from search		
> search()
[1] ".GlobalEnv"         "package:digest"     "package:rflowcyt"
[4] "package:locfit"     "package:akima"      "package:fields"
[7] "package:rfcdmin"    "package:hexbin"     "package:colorspace"
[10] "package:lattice"    "package:splancs"    "package:sp"
[13] "package:grid"       "package:MASS"       "package:KernSmooth"
[16] "package:xtable"     "package:survival"   "package:splines"
[19] "package:methods"    "package:stats"      "package:graphics"
[22] "package:grDevices"  "package:utils"      "package:datasets"
[25] "Autoloads"          "package:base"

I havent yet been able to download R to a windows computer.
Ulrik


On 5/30/06, Ulrik Stervbo <ulriks at ruc.dk> wrote:
> Hi Seth, Hi Nolwenn
>
> I get the same signature as well:
> > library("digest")
> > dataFile <- system.file("fcs/facscan256.fcs", package="rfcdmin")
> > dataFile
> [1] "/usr/lib/R/library/rfcdmin/fcs/facscan256.fcs"
> > digest(dataFile, file=TRUE)
> [1] "bfc603037460c0023569159c7360be0c"
>
> I have not yet tried to load my own datafiles, mainly because I am not
> really sure how. Performing the few examples in the prada package
> works without problems (but then again the readFCS methods are
> probably not the same)
>
> I have been wanting to see if I could reproduce the error on windows.
> However, that as well as testing the new lib may be a few days, since
> other things (laboratory work) are demanding my attention.
>
> I will get back with the results as soon as they get in.
>
> Thanks
> Ulrik
>
> On 5/29/06, Nolwenn LeMeur <nlemeur at fhcrc.org> wrote:
> > Hi Seth, Hi Ulrik,
> >
> > I have verified the md5 hash of the data file on my system and I get the
> > same signature as Seth:
> >
> > > library("digest")
> > > datafile=system.file("fcs/facscan256.fcs",package="rfcdmin")
> > > digest(datafile,file=TRUE)
> > [1] "bfc603037460c0023569159c7360be0c"
> >
> >
> >
> > Nolwenn
> >
> > ***********************************
> > Nolwenn Le Meur, PhD
> > Fred Hutchinson Cancer Research Center
> > Computational Biology
> > 1100 Fairview Ave. N., M2-B876
> > P.O. Box 19024
> > Seattle, WA 98109-1024
> >
> > On Mon, 29 May 2006, Seth Falcon wrote:
> >
> > > Nolwenn LeMeur <nlemeur at fhcrc.org> writes:
> > > > Hi Ulrik,
> > > >
> > > > I was not able to reproduce your error.
> > >
> > > I wonder if everyone is using the exact same dataset?
> > >
> > > > Ulrik Stervbo wrote:
> > > >>Here's what I've done after starting R:
> > > >>
> > > >>
> > > >>>library(rflowcyt)
> > > >>>library(Biobase)
> > > >>>fcs.loc <- system.file("fcs", package = "rfcdmin")
> > > >>>file.location <- paste(fcs.loc, "facscan256.fcs", sep = "/")
> > > >>>FC.FCSRobj <- read.FCS(file.location, UseS3 = TRUE, MY.DEBUG =
> > > >>>TRUE,fcs.type = NULL)
> > >
> > > Here's one way to verify:
> > >
> > > 1. Install the digest package.  From R, do:
> > >
> > >      source("http://bioconductor.org/biocLite.R")
> > >      biocLite("digest")
> > >
> > > 2. Obtain the md5 hash of the data file.  By comparing this output,
> > >    we'll know whether everyone has the same file to begin with.
> > >
> > >      library("rflowcyt")
> > >      library("digest")
> > >
> > >      ## Get the path to the data file:
> > >      dataFile <- system.file("fcs/facscan256.fcs", package="rfcdmin")
> > >      > dataFile
> > >      [1] "/Users/seth/RLIB-2.4/rfcdmin/fcs/facscan256.fcs"
> > >
> > >      ## Compute the md5 hash of the file, you should get the _same_
> > >      ## string as I do here.  If not, we have different files.
> > >      > digest(dataFile, file=TRUE)
> > >      [1] "bfc603037460c0023569159c7360be0c"
> > >
> > > + seth
> > >
> > > _______________________________________________
> > > Bioconductor mailing list
> > > Bioconductor at stat.math.ethz.ch
> > > https://stat.ethz.ch/mailman/listinfo/bioconductor
> > > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
> > >
> >
> > _______________________________________________
> > Bioconductor mailing list
> > Bioconductor at stat.math.ethz.ch
> > https://stat.ethz.ch/mailman/listinfo/bioconductor
> > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
> >
>
>
> --
> Ulrik Stervbo
> Cantianstrasse 23
> 10437 Berlin
> Germany
>
> Home: +49 30 441 12 81
> Handy: +49 178 145 82 31
> Blog: http://ulrikstervbo.blogspot.com
> Mailing-list: http://www.coollist.com/group.cgi?l=ulrik_i_berlin
>


-- 
Ulrik Stervbo
Cantianstrasse 23
10437 Berlin
Germany

Home: +49 30 441 12 81
Handy: +49 178 145 82 31
Blog: http://ulrikstervbo.blogspot.com
Mailing-list: http://www.coollist.com/group.cgi?l=ulrik_i_berlin

More information about the Bioconductor mailing list