Searching without enzyme

[Emmely]

Hi,

We've tried to use MaxQuant with the enzyme setting on 'none', but
this generated the following error message: GetPeptideSet ended
unexpectedly. at a3.a(ji A_0, String[] A_2, String A_3, String A_4,
Boolean A_5, String A_6, Enzyme A_7, Int32 A_8, String[] A_9, Random
A_10)
at kl.a(String A_0, String A_1)
at n5.a(String[] A_0)

We would really like to do the search without an enzyme as the samples
contained peptides that have not been digested, so basically anything
is possible for their sequence. I think the settings for 'none' might
not be correct yet in AndromedaConfig.exe, how should we adapt it to
do a search without an enzyme?

Best regards,
Emmely

[J. Cox]

That should theoretically be possible. Though I must admit that is not really very much tested. Naybe you could try again with 1.1.1.6?

[Emmely]

I've tried it again with 1.1.1.6 but I still get the error. It
occurred almost immediately after the first search was started. I've
made a printscreen of the error message but I don't know how to add it
to this message.

> > Emmely- Hide quoted text -
>
> - Show quoted text -

[J. Cox]

Hi,

I just realized that this is very easy to fix. One should not select 'None' as an enzyme but 'unspecific'. In fact the entry 'None' makes no sense and I will remove it from the default xml file.

Cheers,

Juergen

[Emmely]

I’ve tried to analyse the file with ‘unspecific’ as the enzyme setting
but now the analysis won’t progress beyond the first search. After a
couple of hours CPU gets very low and nothing appears to be happening
anymore. After almost 52 hours of doing the first search I don’t
think that it will continue with the next step so I’ve stopped the
analysis. Interestingly enough, FirstSearcher.exe remains active even
after MaxQuant has been closed. It was not possible to delete the
combined folder because it was still being used by a program,
presumably FirstSearcher.exe since I was able to delete the folder
after I ended the process with the Task Manager. FirstSearcher.exe was
still using CPU. In the MaxQuant google group people are describing
similar problems with long processing times, but I do not have this
problem when I run the analysis with Trypsin/P. The program then
completes the entire process for one file (235Mb, Singlets, smaller
fasta file for the first search) in less than an hour using only 1
thread, which I think is quite acceptable.

> > > - Show quoted text -- Hide quoted text -

[Juergen]

This should be better again with the new version. For unspecific
search it would be good in general to use a very restricted fasta
file, because very many peptide candidates are being generated per
protein. E.g. if you analyze nearly pure samples of one protein it
would make sense to use a fasta file only with that protein sequence
for unspecific search.

[Emmely]

Hi Juergen,

I've tried 1.1.1.14 with an unspecific search over the weekend but it
still gets stuck on the first search with very low CPU.
Today I've tried your suggestion of using a very small database and
the database size is very likely one of the problems. MaxQuant now
gets past the first search, but immediately after the Estimating
Complexity step has finished in Identify I get the following error:

The given key was not present in the dictionary. at
System.ThrowHelper.ThrowKeyNotFoundException()
at System.Collections.Generic.Generic.Dictionary`2.get_Item(TKey key)
at a29a(String A_0, ag4A_1, y1A_2)
at kz.a(ob A_0, ag4A_1, y1A_2, Boolean A_3)
at. Ve.a(ag 4A_0, y1 A_1, vr A_2)
at afm.a(String A_0, vr A_1, Int32 A_2)
at afm.a(String A_0, Int32 A_1)

Do you have any idea what the cause for this error might be?

Emmely

[Juergen]

I found the error for this. Will be fixed in the next version.

[Emmely]

That is great news, thank you!