[BioC] problems reading fcs files / flowcore / read.fcs (David Corne)

[Mike]

Cameron,

It was the white spaces in your FCS file keywords causing the error. I fixed it in 1.22.3, which should appear in BioC in one day or two.
Let me know if you want me send you the tar file directly if you can't wait.

Thanks,

Mike


On 07/20/2012 03:27 AM, Cameron Mackenzie wrote:
> Hi Greg & Mike,
>
> I've tested out the various files with v.1.22.2 and it has solved the problem I was getting with one of the instruments, but not the other. The error message I am getting from read.FCS is "Parameter(s) $DATATYPE not contained in 'x'", so it does sound like a file format issue. I've attached a small example file and any help with it would be really appreciated. I have tested this file and, as of today, it behaves on R 2.14 but not 2.15.
>
> Many thanks,
> Cameron
>
>
> LUX Assure
> Unit 5.3
> Heriot Watt Research Park
> Research Avenue South
> Edinburgh
> EH14 4AP
>   
> Tel: 0131 516 7290
> Fax: 0131 516 7294
> www.luxassure.com
>   
> LUX Assure is a division of LUX Innovate Ltd
>
> "This e-mail is intended for the recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely upon this e-mail. If an addressing or transmission error had mis-directed this e-mail, please notify the author by replying to this e-mail. E-mails are not necessarily secure. LUX Innovate Limited does not accept responsibility for changes made to this message after it was sent. Opinions, conclusions and other information in this message expressed by others or which do not relate to the official business of LUX Innovate Limited shall be understood as neither given nor endorsed by LUX Innovate Limited. It is expressly declared that this e-mail does not constitute nor form part of a contract or unilateral obligation unless the contrary is specifically stated in this e-mail and authorised by LUX Innovate Limited."
> Registered in Scotland No. SC219682
>
> -----Original Message-----
> From: Finak, Greg [mailto:gfinak at fhcrc.org]
> Sent: 13 July 2012 02:04
> To: Kieran O'Neill
> Cc: Mike; David Corne; Cameron Mackenzie; bioconductor at r-project.org; Gottardo, Raphael; Josef Spidlen
> Subject: Re: [BioC] problems reading fcs files / flowcore / read.fcs (David Corne)
>
> Kieran,
> Since it's a bug fix it should get pushed to the main page in the next day or two.
>
> Greg Finak, PhD
> Staff Scientist
> Vaccine and Infectious Disease Division
> Fred Hutchinson Cancer Research Center
> Seattle, WA
> gfinak at fhcrc.org
> (206) 667-3116
>
> On 2012-07-12, at 5:57 PM, "Kieran O'Neill" <koneill at bccrc.ca> wrote:
>
>> Thanks Mike. That sounds like a good solution.
>>
>> Do you have an idea when 1.22.2 will be released?
>>
>> Thanks,
>> Kieran
>>
>> On 12 July 2012 16:21, Mike <wjiang2 at fhcrc.org> wrote:
>>> Hi,All,
>>>
>>> The old version of flowCore didn't take care of double delimiters,
>>> which was valid according to FCS3.0 standard. So we added the patch
>>> to support that. Unfortunately this patch assumed the empty value was
>>> not valid for keyword.(since it will be hard to tell the difference
>>> between one pair of single delimiters with empty value and one
>>> double-delimiter)
>>>
>>> In order to still support the FCS generated by some instruments which
>>> contains empty values , I've added the argument 'emptyValue' to
>>> read.FCS routine.
>>> Default is TRUE, which automatically excludes the possibility of
>>> double delimiters(they are treated as a pair of start and end single
>>> delimiter with an empty value) Otherwise,empty value are not allowed
>>> and double delimiters are parsed as regular character which is part
>>> of value/keyword string.
>>>
>>> The fix will appear in flowCore version 1.22.2.  Let me know if the
>>> fix solve the problem.
>>> Thanks,
>>>
>>> Mike
>>>
>>> On 07/12/2012 08:08 AM, Finak, Greg wrote:
>>>> Thanks, David and Cameron
>>>> I've added Mike Jiang to the CC list, he's the point person on flowCore that will work on fixes.
>>>>
>>>> Best,
>>>>
>>>> Greg Finak, PhD
>>>> Staff Scientist
>>>> Vaccine and Infectious Disease Division Fred Hutchinson Cancer
>>>> Research Center Seattle, WA gfinak at fhcrc.org
>>>> (206) 667-3116
>>>>
>>>> On 2012-07-12, at 7:55 AM, "David Corne" <dwcorne at gmail.com> wrote:
>>>>
>>>>> Dear Greg,
>>>>>
>>>>> Many thanks for that.  I cc Cameron Mackenzie, who is at the
>>>>> coalface trying to deal with this problem.  Yes we expected that
>>>>> this was nonconformist files.
>>>>>
>>>>> Cameron, for specific examples of the most urgent FCS files that
>>>>> you would want to work with PC and the latest R,  I think the thing
>>>>> to do is send the error report, and an associated file if possible
>>>>> (e.g. maybe 0ppm?) direct to Greg, and see if his team can fit that
>>>>> in.
>>>>>
>>>>> VBD
>>>>>
>>>>> On 12 July 2012 15:45, Finak, Greg <gfinak at fhcrc.org> wrote:
>>>>>> Hi, David
>>>>>> In 2.15 there were some changes to the FCS file reading routine to make it compliant with the FCS file standards (it was not compliant prior to 2.15, and led to other errors). Unforunately, the result was that FCS reading was broken (inadvertently) for files from instruments where the manufacturers were not following the standard (apparently there are a lot of manufacturers that don't bother to implement the standard correctly). We're working on correcting these errors as we get bug reports.
>>>>>>
>>>>>> That said, it's next to impossible to help you out if you don't provide us with some more information.. i.e. provide a reproducible example,  sessioninfo(), the error message produced, and even an example of an FCS file where read.FCS fails.
>>>>>>
>>>>>> We want to fix these issues for users, and we understand that having software suddenly break when it was fine one version ago can be frustrating, but complaining into thin air is not really productive for anyone.
>>>>>>
>>>>>> Best,
>>>>>>
>>>>>> Greg Finak, PhD
>>>>>> Staff Scientist
>>>>>> Vaccine and Infectious Disease Division Fred Hutchinson Cancer
>>>>>> Research Center Seattle, WA gfinak at fhcrc.org
>>>>>> (206) 667-3116
>>>>>>
>>>>>>> ------------------------------
>>>>>>>
>>>>>>> Message: 4
>>>>>>> Date: Wed, 11 Jul 2012 11:58:27 +0100
>>>>>>> From: David Corne <dwcorne at gmail.com>
>>>>>>> To: bioconductor at r-project.org
>>>>>>> Subject: [BioC] problems reading fcs files / flowcore / read.fcs
>>>>>>> Message-ID:
>>>>>>>    
>>>>>>> <CAFeka--GMXhzPib9SNWs1oFazAZrvtRSazwPtirTQsVuwnz79g at mail.gmail.c
>>>>>>> om>
>>>>>>> Content-Type: text/plain; charset=ISO-8859-1
>>>>>>>
>>>>>>> Dears All,
>>>>>>>
>>>>>>> I find issues with being able to read FCS files using read.fcs
>>>>>>> from the flowcore package;
>>>>>>>
>>>>>>> So far it seems  FCS files from some instruments can't be read in
>>>>>>> R v
>>>>>>> 2.150 on either linux or windows, but with R V 2.13.0  and 2.14.0
>>>>>>> they can be read in fine.
>>>>>>>
>>>>>>> It would be nice to be able to use the latest version of R - does
>>>>>>> anyone have a fix? Could the flowcore authors maybe release a new
>>>>>>> version where read.fcs works in R 2.15.0 ?
>>>>>>>
>>>>>>> VBD
>>>>>> _______________________________________________
>>>>>> Bioconductor mailing list
>>>>>> Bioconductor at r-project.org
>>>>>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>>>>>> Search the archives:
>>>>>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>>>>>
>>>>> --
>>>>> -------------------------------------------------------------------
>>>>> -------------------
>>>>> twitter: @dwcorne  / quote: "lightning, striking a cup of tea"
>>>>> -------------------------------------------------------------------
>>>>> -------------------
>>>