Re: [spctools-discuss] When to use "Decoy hits to pin down negative distribution" in PeptideProphet

2021-01-20 Thread 'David Shteynberg' via spctools-discuss 
This option changes the way mixture models in the TPP (PeptideProphet) are
generated.  The probabilities are estimated based on the mixture models
learned by PeptideProphet.  When the mixture models change so will the
probabilities.

On Wed, Jan 20, 2021 at 8:44 PM Soroush F  wrote:

> Dear David,
>
> Thanks so much for your time answering my questions.
>
> I gather from your answer that this option is for TPP. If that is the
> case, why does the PeptideProphet probabilities changes when selecting the
> option? Am I missing something?
>
> Many thanks.
> Soroush
>
> On Tuesday, January 19, 2021 at 2:59:56 PM UTC-5 David Shteynberg wrote:
>
>> Dear Soroush,
>>
>> This option makes use of the TPP's semi-supervised learning mode, which
>> can be used in conjunction with semi-parametric mode, at the user's
>> discretion.  Until now, some search engines supported by TPP analysis could
>> only be modeled by PeptideProphet using semi-parametric and semi-supervised
>> that required decoys to be used in the search.  Going forward with the new
>> version of the TPP (pre-release version 6.0.0) this is not going to be
>> the case as we will be providing both types of models for more search
>> engine results.  I hope this answers your question.
>>
>> Cheers,
>> -David
>>
>> On Mon, Jan 18, 2021 at 4:02 PM Soroush F  wrote:
>>
>>> Dear all,
>>>
>>> I was wondering when one should use "Use decoy hits to pin down the
>>> negative distribution" in the Peptide Prophet (run through Petunia)?
>>>
>>> If I want to use Target-Decoy approach to estimate FDR in my analysis,
>>> should I use this box, or it has nothing to do with it?
>>>
>>> Thanks,
>>> Soroush
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "spctools-discuss" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to spctools-discu...@googlegroups.com.
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/spctools-discuss/4592e133-201d-4893-9573-ddee56f1071dn%40googlegroups.com
>>> 
>>> .
>>>
>> --
> You received this message because you are subscribed to the Google Groups
> "spctools-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to spctools-discuss+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/spctools-discuss/a33abb3e-ee23-43f8-aab0-af0c7717b676n%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to spctools-discuss+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/spctools-discuss/CAGJJY%3D_38rzxPZMG2d%2BCmp%3D6E3cuMT1hF92sE4fyVjoU89KjVA%40mail.gmail.com.

Re: [spctools-discuss] When to use "Decoy hits to pin down negative distribution" in PeptideProphet

2021-01-20 Thread Soroush F 
Dear David,

Thanks so much for your time answering my questions.

I gather from your answer that this option is for TPP. If that is the case, 
why does the PeptideProphet probabilities changes when selecting the 
option? Am I missing something?

Many thanks.
Soroush

On Tuesday, January 19, 2021 at 2:59:56 PM UTC-5 David Shteynberg wrote:

> Dear Soroush,
>
> This option makes use of the TPP's semi-supervised learning mode, which 
> can be used in conjunction with semi-parametric mode, at the user's 
> discretion.  Until now, some search engines supported by TPP analysis could 
> only be modeled by PeptideProphet using semi-parametric and semi-supervised 
> that required decoys to be used in the search.  Going forward with the new 
> version of the TPP (pre-release version 6.0.0) this is not going to be 
> the case as we will be providing both types of models for more search 
> engine results.  I hope this answers your question.
>
> Cheers,
> -David
>
> On Mon, Jan 18, 2021 at 4:02 PM Soroush F  wrote:
>
>> Dear all,
>>
>> I was wondering when one should use "Use decoy hits to pin down the 
>> negative distribution" in the Peptide Prophet (run through Petunia)?
>>
>> If I want to use Target-Decoy approach to estimate FDR in my analysis, 
>> should I use this box, or it has nothing to do with it?
>>
>> Thanks,
>> Soroush
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "spctools-discuss" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to spctools-discu...@googlegroups.com.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/spctools-discuss/4592e133-201d-4893-9573-ddee56f1071dn%40googlegroups.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to spctools-discuss+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/spctools-discuss/a33abb3e-ee23-43f8-aab0-af0c7717b676n%40googlegroups.com.

Re: [spctools-discuss] Error generation while converting raw files

2021-01-20 Thread 'David Shteynberg' via spctools-discuss 
I am not sure I know a tool called "ProptideProphet" ;) Perhaps you tried
to run PeptideProphet but it generated no results for you?  When you run
"Analyze Peptides" in the TPP interface, it should create a file called
interact.pep.xml by default, that will contain probabilities among other
information.  Did you run "Analyzed Peptides"?  Were there any messages
reported by the analysis?

On Wed, Jan 20, 2021 at 1:30 AM giangiacomo beretta <
giangiacomo.berett...@gmail.com> wrote:

> Hi David! It works nicely, thank you !
>
> Now I am experiencing another issue of mine :)
>
> I run ProptideProphet on the XML fiel generated by XTandem. I need this
> file presuming that PP will add probabilities to the identified peptides as
> this is requested by SpectraST to generate the corresponding spectral
> library (this is actually my final goal).
>
> However, when I try to do it, SpectraST reports: WARNING -- PEPXML IMPORT:
> Importing a .pep.xml file with no probabilities. PeptideProphet probably
> needs to be run on .pep.xml first.
>
> Maybe probabilities are used for data processing but not appendend to the
> XML output file?
>
> Thanks a lot in advance!
>
> G
>
> Il giorno martedì 19 gennaio 2021 alle 20:53:57 UTC+1 David Shteynberg ha
> scritto:
>
>> Dear Giangiacomo,
>>
>> Thanks for trying the TPP and reporting the problem.  TPP uses the
>> proteowizard's msconvert tools for this step.  You can remedy the problem
>> by either upgrading to a newer version of proteowizard's msconvert tool or
>> use  --ignoreUnknownInstrumentError option with your current version.  On
>> the generate mzML page you can specify this option in the"Enter
>> additional options to pass directly to the command-line" text box, just
>> enter the text  --ignoreUnknownInstrumentError
>>
>> Hope it works!
>>
>> Cheers,
>> -David
>>
>>
>> On Tue, Jan 19, 2021 at 8:10 AM giangiacomo beretta <
>> giangiacom...@gmail.com> wrote:
>>
>>> Hi, when I try to convert raw files, the conversion tool stops
>>> immediately reporting the following error:
>>>
>>> Reader_Thermo::fillInMetadata] unable to parse instrument model; please
>>> report this error to the ProteoWizard developers with this information:
>>> model(Orbitrap Eclipse) name(Orbitrap Eclipse); if want to convert the file
>>> anyway, use the ignoreUnknownInstrumentError flag
>>>
>>> It appears that the system is not recognizing the instrument model that
>>> produced the raw files. Is there any option to overcome this issue?
>>>
>>> Actually I have converted the same files with ProteoWizard without
>>> troubles.
>>>
>>> Thanks in advance!
>>>
>>> G
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "spctools-discuss" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to spctools-discu...@googlegroups.com.
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/spctools-discuss/2134bb9e-bb7e-4f52-9afc-bf59b75f21aen%40googlegroups.com
>>> 
>>> .
>>>
>> --
> You received this message because you are subscribed to the Google Groups
> "spctools-discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to spctools-discuss+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/spctools-discuss/95da30b4-f9d1-49cc-896e-fe734ba0cac4n%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to spctools-discuss+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/spctools-discuss/CAGJJY%3D9HEf8jtFUOh0yE1n3bT_%2BFz3VKvqd48oJkaHjJHgDMVg%40mail.gmail.com.

Re: [spctools-discuss] Error generation while converting raw files

2021-01-20 Thread giangiacomo beretta 
Hi David! It works nicely, thank you ! 

Now I am experiencing another issue of mine :)

I run ProptideProphet on the XML fiel generated by XTandem. I need this 
file presuming that PP will add probabilities to the identified peptides as 
this is requested by SpectraST to generate the corresponding spectral 
library (this is actually my final goal).

However, when I try to do it, SpectraST reports: WARNING -- PEPXML IMPORT: 
Importing a .pep.xml file with no probabilities. PeptideProphet probably 
needs to be run on .pep.xml first.

Maybe probabilities are used for data processing but not appendend to the 
XML output file?

Thanks a lot in advance!

G

Il giorno martedì 19 gennaio 2021 alle 20:53:57 UTC+1 David Shteynberg ha 
scritto:

> Dear Giangiacomo,
>
> Thanks for trying the TPP and reporting the problem.  TPP uses the 
> proteowizard's msconvert tools for this step.  You can remedy the problem 
> by either upgrading to a newer version of proteowizard's msconvert tool or 
> use  --ignoreUnknownInstrumentError option with your current version.  On 
> the generate mzML page you can specify this option in the"Enter 
> additional options to pass directly to the command-line" text box, just 
> enter the text  --ignoreUnknownInstrumentError 
>
> Hope it works!
>
> Cheers,
> -David   
>
>
> On Tue, Jan 19, 2021 at 8:10 AM giangiacomo beretta <
> giangiacom...@gmail.com> wrote:
>
>> Hi, when I try to convert raw files, the conversion tool stops 
>> immediately reporting the following error:
>>
>> Reader_Thermo::fillInMetadata] unable to parse instrument model; please 
>> report this error to the ProteoWizard developers with this information: 
>> model(Orbitrap Eclipse) name(Orbitrap Eclipse); if want to convert the file 
>> anyway, use the ignoreUnknownInstrumentError flag 
>>
>> It appears that the system is not recognizing the instrument model that 
>> produced the raw files. Is there any option to overcome this issue?
>>
>> Actually I have converted the same files with ProteoWizard without 
>> troubles.
>>
>> Thanks in advance!
>>
>> G
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "spctools-discuss" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to spctools-discu...@googlegroups.com.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/spctools-discuss/2134bb9e-bb7e-4f52-9afc-bf59b75f21aen%40googlegroups.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to spctools-discuss+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/spctools-discuss/95da30b4-f9d1-49cc-896e-fe734ba0cac4n%40googlegroups.com.