FCS files and software compatability

Robert C. Leif rleif at rleif.com
Tue Nov 1 19:42:44 EST 2005


Simon, Ryan, et al.

Several other groups are working on this problem. These include the
Laboratory Digital Imaging Project (LDIP) Data Exchange Specification
(http://www.ldip.org/), extensions to the Digital Imaging and Communications
in Medicine (DICOM,   <http://medical.nema.org/dicom/2004.html>
http://medical.nema.org/dicom/2004.html, the Open Microscopy Environment
(OME) ( <http://www.openmicroscopy.org/> http://www.openmicroscopy.org/),
Ryan's group, and my work on CytometryML
(http://www.newportinstruments.com/cytometryml/cytometryml.htm).

A group of which I am a member has started to extend DICOM for clinical
digital microscopy.  I intend to extend this to include flow cytometry.
Cytometrists, pathologists, and most importantly the patients will have a
big problem.  Each society or group is creating its own stovepipe solution
or standard.  If this is not corrected, the interoperability of informatics
systems for cytometry and pathology will be gravely impaired.  

Since I could not convince ISAC to take part in DICOM, I have translated a
significant part of the relevant DICOM datatypes into XML schema and plan to
extend these schemas with some of the datatypes from the OME schema.
Schemas to describe both images and flow data have been included in the
latest version of CytometryML, which I hope to post on the Newport
Instruments website in the near future.  LDIP is working on a XML Resource
Description Framework (RDF) compilation that will include the relevant DICOM
methods.  The OME has some tools that should be of interest to users of
digital microscopy. 

I will be speaking at the SPIE BiOS meeting
(http://spie.org/app/conferences/index.cfm) , which takes place on 21 to 26
January in San Jose and hope to present at the 2006 ISAC Congress. 

Bob Leif



-----Original Message-----
From: Ryan Brinkman [mailto:rbrinkman at bccrc.ca] 
Sent: Monday, October 31, 2005 11:22 AM
To: cyto-inbox
Subject: RE: FCS files and software compatability



Simon,


Along with several collaborators we are currently working on developing

specifications that will hopefully aid the analysis of data across

different software such as you are describing. Our goal is to facilitate

the interchange of flow cytometric analyses between software packages,

investigators and laboratories, including data exchange standards for

analysis (especially gating) and experimental meta data. However that

doesn't help you today. 


What should help you is that one of the co-investigators on our project

is Robert Gentleman, a co-developer of R, who is leading the work on

rflowcyt. While but we are also working on example implementations in

Java (with help from Perry Haaland's group at BD and Adam Treister's

group TreeStar), the R package is the most developed at the moment on

the analysis side. Based on Robert's work (and Tony Rossini before him)

you don't need to convert your data to text to get it into R, rflowcyt

reads FCS files directly into R data objects. Robert released an update

to rflowcyt a couple of weeks ago that you can access as part of the

BioConductor package (http://www.bioconductor.org/). This release fixes

some issues and adds further functionality. There is ample documentation

for R that you should look at first

(http://cran.r-project.org/manuals.html,

http://cran.r-project.org/other-docs.html) along with most importantly

the 61-page manual on rflowcyt itself

(http://www.bioconductor.org/docs/vignettes.html) which nicely explains

how to do this. After you have looked at all that documentation you

could also get in touch with Nolwenn LeMeur (nlemeur at fhcrc.org), who has

done much of the recent work on rflowcyt, as she is also in Seattle.


Cheers,

Ryan


Ryan Brinkman, PhD

Senior Scientist, Terry Fox Laboratory

BC Cancer Research Centre &

Assistant Professor, Medical Genetics, UBC

675 West 10th Avenue

Vancouver, BC V5Z 1L3

Tel: (604) 675-8132

http://www.bccrc.ca/tfl


-----Original Message-----

From: Mr Simon Corrie [mailto:s369338 at student.uq.edu.au] 

Sent: Sunday, October 30, 2005 9:18 AM

To: cyto-inbox

Subject: FCS files and software compatability


Hi folks


Im doing a short research project in Seattle and as such will have 

data from three different machines (Moflo, LSR2, Beckman MPLFC500) and 

three different software (mainly analysis) packages (FACSDiva, Summit, 

CPX).


I am finding it difficult to analyse my data across softwares - 

generally the problem seems to relate to the high resolution of the 

FACSDiva package, such that opening data in summit or CPX gives me 

data in the very low quadrant (log) areas. To solve the problem I 

would like to find a convenient method to convert my FCS files from 

various sources into .txt files so that I can use Matlab or R to 

analyse the data.


So finally my questions are:

1) Can someone suggest a program for the batchwise conversion of FCS 

to text? (I have tried ldata, but the FCS2.0 and 3.0 files do not 

convert)


2) Can someone give me a reference to help me learn how to set up my 

statistical analysis in R? ie maybe the most recent guide to rflowcyt 

if thats what people are using or an easier method? (I can code in 

matlab.....R frankly looks scary with a lack of reference material..)


Any ideas would be much appreciated


Regards


Simon Corrie

PhD Student

Nanotechnology and Biomaterials Centre

University of Queensland

St Lucia 4072 QLD Australia

-------------- next part --------------
HTML attachment scrubbed and removed


More information about the Cytometry mailing list