digital data transformation revert
This message: [ Message body ] [ More options ]
Related messages: [ Next message ] [ Previous message ] [ In reply
to ]
From: Julia Whynot <whynotj@[EMAIL PROTECTED]
>
Date: Thu Mar 16 2006 - 17:43:24 EST
While we are on the subject. I have listened to this discussion and
have played around a bit with the transformation in flowjo using data
acquired on our FACScanto Diva software.
How does one revert back to the original view without having to bring
the file back in?
On Thursday, March 16, 2006, at 02:06 AM, A.J. Rossini wrote:
> On 3/13/06, Mario Roederer <roederer@[EMAIL PROTECTED]
> wrote:
>> I am a firm believer in the use of these transformations--in the
>> multicolor> world,
> they have revolutionized our ability to look at the data. (Note that>
> the transforms in
> no way affect statistics or gating--they ONLY affect the> visual
> representation of the
> data). The point of these transforms is to> make the data more easily
> conveyed to
> readers--something that a simple log> scale (which is itself a
> transformation!) no longer
> does adequately.
> I'm going to nitpick a bit -- while the general overall
> conclusionsMario draws are
> correct, there are a few subtle points. If you don'tget what I'm
> saying, believe Mario,
> it's simpler that way. Read onfor the fine details:
> Transformations actually do affect the statistical
> analysis/conclusions.
> data != transformed data.
> Conclusions drawn on transformed data do not have to be the same as
> onraw data (a
> difference on a log scale using a particular statisticisn't
> necessarily different using
> the raw data with the samestatistic).
> Why transform?
> To pull the data back to a state of nature where we feel
> comfortablewith decision making.
> To pull the data back to a state of nature where certain
> probabilisticassumptions are
> met, allowing for "believable" statistical estimationand testing (and
> opposed to
> unbelieveable statistical muck)
> To focus on particular ranges or features of the data.
> All are valid.
> The detail is just "comparison on the transformed scale isn't the
> sameas comparison on
> the original scale". The point is that it's ratherim****tant to do,
> both to match up
> with historical scientificapproaches as well as to match up with
> common practices
> forcomparability within the cytometric field.
>>> The Data Presentation Standards Committee, formed by ISAC, will take
>>> the>
> responsibility to contact the editors of the Journal with which there
> are> problems. Our
> goal is to educate not only the reviewers, but im****tantly> the
> Editors--so that they can
> be made aware that any complaints that> reviewers might have regarding
> this visualization
> are baseless.
> This is im****tant! Visualizations, done well, are critical
> formultidimensional data.
> best,-tony
> (Novartis Pharma AG / University of Wa****ngton)
> blindglobe@[EMAIL PROTECTED]
Switzerland."Commit early,commit often,
> and commit in a
> repository from which we can easilyroll-back your mistakes" (AJR,
> 4Jan05).
>
Julia Whynot
Rockefeller University
Investigative Dermatology
212-327-7581
Received on Mon Mar 20 20:38:00 2006
This message: [ Message body ]
Next message: Moody, Pamela: "Meeting Next Week at Cold Spring Harbor
Laboratories"
Previous message: Derek Davies: "UK position"
In reply to: A.J. Rossini: "Re: Re digital data"
Contem****ary messages sorted: [ By Date ] [ By Thread ] [ By Subject ]
[ By Author ] [ By messages with attachments ]
This archive was generated by hypermail 2.1.8 : Tue Mar 21 2006 -
03:12:02 EST
|
