Magnus,

The expectation value calculation does not depend on the batch size at all.
 It's calculated for each spectrum query and is only a function of the
spectrum and its peptide search space (where search space is defined by the
database and the search parameters).

- Jimmy


On Tue, Mar 4, 2014 at 2:12 AM, [email protected] <
[email protected]> wrote:

> Hi Jimmy,
>
> Many thanks - the "spectrum_batch_size" trick worked! Are the results
> (e.g. the expectation values) dependent on the batch size?
>
>
> Cheers,
>
> Magnus
>
>
> On Monday, 3 March 2014 19:05:21 UTC+1, Jimmy Eng wrote:
>
>> Mangus,
>>
>> There's something broken with our latest Comet 2013.02 rev. 2 release
>> under Windows.  We attempted to address large memory search performance,
>> which appeared to work fine in our limited testing, but we definitely broke
>> something else in the process and are actively working on addressing this
>> issue now.
>>
>> The current suggested work-around would be:
>> a).  make use of the "spectrum_batch_size" parameter; set it to some
>> value, say 3000 or 5000 to allow you to process your data by searching
>> smaller batches of spectra at a time
>> b).  revert the Comet binary to the previous 2013.02 rev. 1 release.
>>
>> - Jimmy
>>
>>
>> On Mon, Mar 3, 2014 at 9:46 AM, [email protected] <
>> [email protected]> wrote:
>>
>>> Dear TPP developers and users,
>>>
>>> I took the the new Comet in TPP 4.7 (POLAR VORTEX) out for spin, but
>>> keep getting errors like "Error - 
>>> CometMemMgr::CometMemAlloc(pdTmpFastXcorrData[1374])"
>>> and "Operation not permitted" when searching an mzXML made from some 5 GB
>>> Orbitrap E. coli data. This mzXML file works fine with both X!Tandem and
>>> SpectraST. To get a parameter file, I created one with "comet -p".
>>>
>>> When I search a tiny mzXML file with one MS1 and three MS2 spectra I
>>> created myself, it works, and I get reasonable results. But I have tried it
>>> with reasonably small ion trap datasets of <100 MB, I get the errors as for
>>> the 5 GB Orbi data.
>>>
>>> Any suggestions? Should I allocate more memory to comet somewhere?
>>>
>>>
>>> Cheers,
>>>
>>> Magnus
>>>
>>> --
>>> 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 [email protected].
>>> To post to this group, send email to [email protected].
>>>
>>> Visit this group at http://groups.google.com/group/spctools-discuss.
>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>
>>
>>  --
> 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 [email protected].
> To post to this group, send email to [email protected].
> Visit this group at http://groups.google.com/group/spctools-discuss.
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/spctools-discuss.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to