Hello Ping,
Yes, please make sure you run xinteract with the -p0 (zero, not "O")
option.  QualScore uses those to model the "bad" spectra.
--Luis

On Thu, Apr 16, 2009 at 10:43 AM, Ping <[email protected]> wrote:

>
> Luis,
>
> I finally got my PepXMLViewer working. I opened the file, and there is
> no entries with exact zero
> probability. But there are tons of entries with ~0.05 probability.
> Does that matter? Or QualScore
> only considers that zero probability as bad spectra?
>
> The data is LTQ data, and first is searched under X!tandem. The result
> I used is from TPP.
>
> Thanks again!
>
> Ping
>
> On Apr 15, 3:42 pm, Luis Mendoza <[email protected]> wrote:
> > Hi Ping,
> > By "the same results", do you mean that all of the program output is
> > identical? (e.g. the number of spectra used, etc)  Or just that you also
> get
> > an unusable training set?  I would first make sure that you are setting
> the
> > -p0 option successfully by opening the PeptideProphet results in the
> PepXML
> > Viewer, and look for entries with zero probability; there should be many.
> > You can also look at the following thread for some other hints:
> http://groups.google.com/group/spctools-discuss/browse_thread/thread/.<http://groups.google.com/group/spctools-discuss/browse_thread/thread/>
> ..
> >
> > Otherwise the software does suggest that it cannot find any "bad" spectra
> to
> > derive a training set.  What kind of instrument did you use to acquire
> your
> > data?  Did you do any pre-processing before searching?
> > --Luis
> >
>  > On Wed, Apr 15, 2009 at 1:43 PM, Ping <[email protected]> wrote:
> >
> > > Hi Luis,
> >
> > > Thanks a lot for your response.
> >
> > > I used the x!Tandem results, all.pep.xml  + original mzXML files as
> > > input. But I got following error.
> >
> > > Features calculated for:  24780 (1171916 ms)
> > > Training set:  12699/0 (good/bad)
> > > Unassigned spectra:  2611
> > > ERROR: Training set distribution unusable.
> >
> > > I searched the post in the group, and used -p0 option, but still I got
> > > the same results. Does that mean my original mzXML has too little low
> > > quality spectral data?
> >
> > > Thanks again,
> >
> > > Ping
> >
> > > On Apr 15, 11:44 am, Luis Mendoza <[email protected]> wrote:
> > > > Hello Ping,
> > > > Qualscore uses a pepXML file with *all* PeptideProphet probabilities
> > > (i.e.
> > > > P>=0.0).  As such, you will first need to search your data and run it
> > > > through PeptideProphet before you can use QualScore.  You can find
> more
> > > > information about the algorithm in the publication listed in the
> README
> > > > file:
> > >http://sashimi.svn.sourceforge.net/viewvc/sashimi/trunk/qualscore/REA.
> ..
> >
> > > > Best of luck, and thanks for trying out our software.
> > > > --Luis
> >
> > >  > On Wed, Apr 15, 2009 at 10:15 AM, Ping <[email protected]> wrote:
> >
> > > > > Hi All,
> >
> > > > > I just downloaded the qualscore from sashimi, and installed it on
> my
> > > > > Ubuntu machine. The command line should be:
> >
> > > > > java -jar qualscore_v1.0_2.jar someXML file
> >
> > > > > I am not sure what this XML suppose to be look like? Can anyone
> send
> > > > > me an example of it?
> >
> > > > > I am a newbie in this area. So please forgive me if the question is
> > > > > too simple.
> >
> > > > > Thanks!
> >
> > > > > Ping
> >
>

--~--~---------~--~----~------------~-------~--~----~
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