Your best bet is to try the code from trunk SVN.  Have you tried it in
NONPARAM mdoe?

-David

On Wed, Feb 2, 2011 at 12:52 PM, Andreas Quandt
<[email protected]> wrote:
> hey jimmy and matt,
> i tried what jimmy suggested but are stuck again.... :-(
> i added the 'missing' scores as jimmy suggested:
>         <search_score name="deltacn" value="0.22324939" />
>         <search_score name="xcorr" value="0.96492940" />
>         <search_score name="deltacnstar" value="0.000"/>
>         <search_score name="spscore" value="0.0000"/>
>         <search_score name="sprank" value="1"/>
> and get following output:
> running: "/cluster/apps/tpp/4.4.1/bin/PeptideProphetParser
> 'interact.pep.xml' DECOY=DECOY_ ACCMASS DECOYPROBS NONPARAM LEAVE PI
> INSTRWARN"
> using Accurate Mass Bins
> using pI
> Using no error on different instrument types.
> Using Decoy Label "DECOY_".
> Decoy Probabilities will be reported.
> Using non-parametric distributions
>  (SEQUEST) (leaving deltacn* entries)
> adding Accurate Mass mixture distr
> adding pI mixture distr
> init with SEQUEST trypsin
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.00210.00210.4. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.00234.00234.3. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.00668.00668.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.01888.01888.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.01912.01912.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.01913.01913.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.02423.02423.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.02440.02440.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.02441.02441.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.03150.03150.3. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.03171.03171.3. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.04499.04499.3. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.04599.04599.4. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.06133.06133.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.06134.06134.2. This spectrum will be excluded from the model.
> ERROR: Discriminant value computed as nan, for spectrum
> B08-02057.07676.07676.4. This spectrum will be excluded from the model.
> MS Instrument info: Manufacturer: NA, Model: NA, Ionization: NA, Analyzer:
> UNKNOWN, Detector: UNKNOWN
>  PeptideProphet  (TPP v4.4 VUVUZELA rev 1, Build 201102011840 (linux))
> AKeller@ISB
>  read in 0 1+, 5 2+, 416 3+, 26 4+, 2 5+, 1 6+, and 0 7+ spectra.
> Initialising statistical models ...
> Found 238 Decoys, and 212 Non-Decoys
> Iterations: .PeptideProphetParser:
> Validation/InterProphet/InterProphetParser/KDModel.cxx:143: bool
> KDModel::replaceProb(int, double): Assertion `idx < posprobs_->size()'
> failed.
> command "/cluster/apps/tpp/4.4.1/bin/PeptideProphetParser 'interact.pep.xml'
> DECOY=DECOY_ ACCMASS DECOYPROBS NONPARAM LEAVE PI INSTRWARN" exited with
> non-zero exit code: 6
> QUIT - the job is incomplete
> the errors are coursed by negative values for the xcorr. hence i 'turned'
> them into positive values. however this only causes another error:
> command completed in 1 sec
> running: "/cluster/apps/tpp/4.4.1/bin/PeptideProphetParser
> 'interact.pep.xml' DECOY=DECOY_ ACCMASS DECOYPROBS NONPARAM LEAVE PI
> INSTRWARN"
> using Accurate Mass Bins
> using pI
> Using no error on different instrument types.
> Using Decoy Label "DECOY_".
> Decoy Probabilities will be reported.
> Using non-parametric distributions
>  (SEQUEST) (leaving deltacn* entries)
> adding Accurate Mass mixture distr
> adding pI mixture distr
> init with SEQUEST trypsin
> command "/cluster/apps/tpp/4.4.1/bin/PeptideProphetParser 'interact.pep.xml'
> DECOY=DECOY_ ACCMASS DECOYPROBS NONPARAM LEAVE PI INSTRWARN" exited with
> non-zero exit code: 11
> QUIT - the job is incomplete
>
> as i do not know what these error codes mean, i am now stuck :-(
> does one of you guys have a suggestion how to continue?
> cheers,
> andreas
> On Wed, Feb 2, 2011 at 6:49 PM, Andreas Quandt <[email protected]>
> wrote:
>>
>> Hey Jimmy,
>>
>> I tried the renaming but of course not the adding of the 'missing'
>> scoring values.
>> I report back to you guys as soon as I have tested it :-)
>>
>> Cheers,
>> Andreas
>>
>>
>>
>> On Feb 2, 2011, at 5:51 PM, Jimmy Eng <[email protected]> wrote:
>>
>> > Andreas,
>> >
>> > I didn't say it was going to be easy! :)
>> >
>> > Here's an example Sequest pep.xml file:
>> > http://proteomicsresource.washington.edu/dist/sequest.pep.xml
>> >
>> > As Matt suggested, the search_score names will need to change
>> > ("xcorr_score" to "xcorr" and "delta_cn" to "deltacn").  Additionally,
>> > you'll probably want to add "deltacnstar", "spscore", and "sprank"
>> > score entries.  Use 0.0, 0.0, and 1 respectively for the values for
>> > these additional scores.
>> >
>> > - Jimmy
>> >
>> > On Wed, Feb 2, 2011 at 8:42 AM, Matthew Chambers
>> > <[email protected]> wrote:
>> >> Did you try renaming the score names? I expect that there is a 1:1
>> >> relationship between the search engine name and the score names
>> >> (correct me
>> >> if I'm wrong TPP folks). If you need a program to do the "replace all"
>> >> you
>> >> can try Notepad++ or Gvim.
>> >>
>> >> -Matt
>> >>
>> >>
>> >> On 2/2/2011 10:22 AM, Andreas Quandt wrote:
>> >>>
>> >>> sure :-)
>> >>>
>> >>> originally the spectrum query match looks like
>> >>>
>> >>> <spectrum_query spectrum="B08-02057.mzXML.00234.00234.3"
>> >>>
>> >>> but after it did not work. i tried the B08-02057.00234.00234.3
>> >>> instead:
>> >>>
>> >>>
>> >>> <spectrum_query spectrum="B08-02057.00234.00234.3" start_scan="234"
>> >>> end_scan="234"
>> >>> precursor_neutral_mass="1332.6912" assumed_charge="3" index="4">
>> >>> <search_result>
>> >>> <search_hit hit_rank="2" peptide="GAFALTLIGLLLM" peptide_prev_aa="R"
>> >>> peptide_next_aa="-"
>> >>> protein="DECOY_Q8TE58" num_tot_proteins="1"
>> >>> calc_neutral_pep_mass="1332.7074" massdiff="-0.0162"
>> >>> num_tol_term="2" num_missed_cleavages="0"  is_rejected="0"
>> >>> protein_descr="\ID=ATS15_HUMAN \MODRES=
>> >>> \VARIANT=(623|N|S)(770|Q|R)(878|C|G) \NCBITAXID=9606 \DE=A disintegrin
>> >>> and
>> >>> metalloproteinase with
>> >>> thrombospondin motifs 15 (ADAMTS-15) (ADAM-TS 15) (ADAM-TS15)">
>> >>> <search_score name="delta_cn" value="0.00000000" />
>> >>> <search_score name="xcorr_score" value="-0.04533549" />
>> >>> </search_hit>
>> >>> </search_result>
>> >>> </spectrum_query>
>> >>> <spectrum_query spectrum="B08-02057.00244.00244.5" start_scan="244"
>> >>> end_scan="244"
>> >>> precursor_neutral_mass="2591.2581" assumed_charge="5" index="5">
>> >>> <search_result>
>> >>> <search_hit hit_rank="1" peptide="HMLDVLLLFIAVIVALPVLAELGM"
>> >>> peptide_prev_aa="R" peptide_next_aa="-"
>> >>> protein="DECOY_P10635" num_tot_proteins="1"
>> >>> calc_neutral_pep_mass="2591.2922" massdiff="-0.0342"
>> >>> num_tol_term="2" num_missed_cleavages="0"  is_rejected="0"
>> >>> protein_descr="\ID=CP2D6_HUMAN \MODRES=
>> >>>
>> >>>
>> >>> \VARIANT=(11|V|M)(26|R|H)(28|R|C)(34|P|S)(42|G|R)(85|A|V)(91|L|M)(94|H|R)(107|T|I)(120|F|I)(155|E|K)(169|G|R)(212|G|E)(231|L|P)(237|A|S)(281|.|)(296|R|C)(297|I|">
>> >>> <search_score name="delta_cn" value="0.00000000" />
>> >>> <search_score name="xcorr_score" value="0.37809423" />
>> >>> </search_hit>
>> >>> </search_result>
>> >>> </spectrum_query>
>> >>> <spectrum_query spectrum="B08-02057.00246.00246.4" start_scan="246"
>> >>> end_scan="246"
>> >>> precursor_neutral_mass="2007.0232" assumed_charge="4" index="6">
>> >>> <search_result>
>> >>>
>> >>> (to the original sequest, the difference is also that the scores are
>> >>> named: xcorr and deltacn...)
>> >>>
>> >>> --andreas
>> >>>
>> >>> On Wed, Feb 2, 2011 at 5:04 PM, Matthew Chambers
>> >>> <[email protected]
>> >>> <mailto:[email protected]>> wrote:
>> >>>
>> >>>    Can you paste a few of the spectrum_query elements? It seems like
>> >>> you
>> >>> got it to process properly
>> >>>    as SEQUEST, so there's something wrong further down.
>> >>>
>> >>>    -Matt
>> >>
>> >> --
>> >> 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.
>> >
>
> --
> 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