Shaun, The latest TPP supports CRUX output.  It has been awhile since
I implemented this support but it appears that the Crux function is
very similar to TPP sequest discriminant but only using Xcorr and
delta Cn.

Pumuta, I have not yet tried the Q-ranker code, but the PeptideProphet
paper you site is the original from 9 years.  Since then there have
been a number of improvements to the algorithm.  Perhaps the paper you
should be citing is the iProphet paper that show using the TPP on the
the results of a single search engine or on combining multiple search
engines the paper can be found here
http://mcponline.org/content/early/2011/08/29/mcp.M111.007690.abstract?sid=a82bdd5e-a70e-464a-a5e5-292c8a00a3f7.
 Perhaps we will integrate the Q-ranker algorithm into the TPP in the
future, in fact I am currently trying to install this software on my
systems.
Until then however, I will offer that instead of trying to Jerry-rig
tools not yet integrated with the TPP to the ProteinProphet algorithm
is awkward, likely to cause you more potential problems and headaches,
than if you simply:

1. download and install the latest TPP version from here:
http://sourceforge.net/projects/sashimi/files/Trans-Proteomic%20Pipeline%20%28TPP%29/TPP%20v4.5%20%28rapture%29%20rev%200/

2. run your Crux data through the pipeline

Good luck,
-David


On Mon, Oct 3, 2011 at 11:55 PM, Putuma Gqamana
<[email protected]> wrote:
> Peptide Prophet and post-search algorithms like Q-ranker, Percolator
> and Q value all treat SEQUEST output data and create discriminants
> from it, i.e., Prophet discriminant has 4 SEQUEST terms compared to
> Percolator's 17 terms of which 10 are SEQUEST based and Q-ranker's 34
> terms. So I take it that Peptide Prophet cannot treat CRUX outputs
> which are already post-search statistically treated outputs, but it
> definitely treats SEQUEST outputs. Of course, you already know that
> Q-ranker > Percolator > Prophet, so what you are trying to do is kinda
> awkward or retrogressive here. You'd rather subject a CRUX Qranker
> pepXML output or percolator.pepXML to Protein Prophet instead.
>
> On Mon, Oct 3, 2011 at 12:12 PM, naturofix <[email protected]> wrote:
>> Does anyone know how PeptideProphet generates the discriminant score
>> when it gets an input from Crux, for instance the
>> qranker.target.pep.xml output. These files contain a q-ranker_score
>> and a q_ranker_q_value as well as delta_cn and Xcorr_score.
>>
>> Which of these statistics are used to generate the probability score
>> in the interact.pep.xml output.
>> Where can I find a formula for this calculation. Does the same hold
>> for percolator?
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups 
>> "spctools-discuss" group.
>> To post to this group, send email to [email protected].
>> To unsubscribe from this group, send email to 
>> [email protected].
>> For more options, visit this group at 
>> http://groups.google.com/group/spctools-discuss?hl=en.
>>
>>
>
>
>
> --
> Putuma P. Gqamana, Ph.D. - Wellcome Trust - CIDRI Postdoctoral Research Fellow
> University of Cape Town Faculty of Health Sciences
> c/o Blackburn Group - Division of Medical Biochemistry
> N3.06.4 Wernher and Beit Building North
> Anzio Road,
> Observatory, 7925
>
> Email : [email protected], [email protected], [email protected]
> Phone: ++27-21-406-6453 (w)
> Phone: ++27-73-764-7825 (cell)
> Skype: Putuma.Gqamana
>
> --
> You received this message because you are subscribed to the Google Groups 
> "spctools-discuss" group.
> To post to this group, send email to [email protected].
> To unsubscribe from this group, send email to 
> [email protected].
> For more options, visit this group at 
> http://groups.google.com/group/spctools-discuss?hl=en.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"spctools-discuss" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/spctools-discuss?hl=en.

Reply via email to