There is a bug to be reported and a question

[Fengchao Yu]

Dear all,

I want to report a bug:

If I set "verbosity=40" in parameter file during "search-for-xlinks", it seems that the software will not work. It won't produce any result. It's strange!

Also, there is a question when I use the "search-for-xlinks":

How can I set the option "--xlink-prevents-cleavage" so that the xlink won't prevents any amino acid from cleavaging?

Thx

Fengchao

[William S Noble]

On Fri, Feb 1, 2013 at 1:01 AM, Fengchao Yu <fcy...@gmail.com> wrote:

Dear all,

I want to report a bug:

If I set "verbosity=40" in parameter file during "search-for-xlinks", it seems that the software will not work. It won't produce any result. It's strange!

 Hi Fengchao,

I agree that this sounds strange indeed.  Sean will look into this tonight and get back to you with a fix.

 

Also, there is a question when I use the "search-for-xlinks":

How can I set the option "--xlink-prevents-cleavage" so that the xlink won't prevents any amino acid from cleavaging?

At the moment, the only option is to explicitly list all twenty amino acids (i.e., "--xlink-prevents-cleavage ACDEFGHIJKLMNPQRSTVWY").  In the next release, we will add a wildcard character to make this easier to do.

Cheers.

Bill

Thx

Fengchao

--
You received this message because you are subscribed to the Google Groups "crux-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to crux-users+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Fengchao Yu]

Thanks,

By the way, there is another bug: 

During the search, there may be a warning :"Ignoring peptides with ambiguous amino acids (B,X,Z).

As we all know, "B" means Aspartic acid or Asparagine; X means any; Z means Glutamic acid or Glutamine. So, I wonder if the next release will fix this problem, or there is any solution to this.

Thanks again,

Fengchao

[Fengchao]

Hi Sean,

Thank you for your effort. I re-ran crux with different kinds of data. Finally, I found where the problem is. It's not about the verbosity. It's about the MzXML version crux supports. It seems that crux don't support MzXML 3.0 that use zlib to compress the binary data. The old version of mzXML that don't compress the binary data with zlib is fully supported by crux. I put a website below to help you understand the difference of mzXML 3.2:
http://www.proteomecenter.org/ASMS/Pedrioli_ASMS_2006.pdf

By the way, the bug I reported many days ago about the "segmentation fault (core dump)" is still unsolved (see the google groups for details). I think this is a serious problem.

Thanks again,

Fengchao

Sean McIlwain

Tuesday, February 05, 2013 1:10 PM

Fengchao,

Could you send me your files?  I'm having trouble reproducing this bug.

Thanks,

Sean

William S Noble

Tuesday, February 05, 2013 1:33 AM

Fengchao Yu

Friday, February 01, 2013 5:01 PM

Dear all,

I want to report a bug:

If I set "verbosity=40" in parameter file during "search-for-xlinks", it seems that the software will not work. It won't produce any result. It's strange!

Also, there is a question when I use the "search-for-xlinks":

How can I set the option "--xlink-prevents-cleavage" so that the xlink won't prevents any amino acid from cleavaging?

[Sean McIlwain]

Fengchao,

On Feb 5, 2013, at 3:14 AM, Fengchao wrote:

Hi Sean,

Thank you for your effort. I re-ran crux with different kinds of data. Finally, I found where the problem is. It's not about the verbosity. It's about the MzXML version crux supports. It seems that crux don't support MzXML 3.0 that use zlib to compress the binary data. The old version of mzXML that don't compress the binary data with zlib is fully supported by crux. I put a website below to help you understand the difference of mzXML 3.2:
http://www.proteomecenter.org/ASMS/Pedrioli_ASMS_2006.pdf

Which spectrum parser are you using?  MSToolkit (spectrum-parser=mstoolkit) and Proteowizard (spectrum-parser=pwiz) both parse mzXML.  Recently,  we posted a fix to the crux-proteowizard interface that could fix the problem (for both MGF and mzXML in proteowizard).  This fix should be in the next release, but could you try that to see if it fixes the problem for you?

https://groups.google.com/d/topic/crux-users/ftkHjE94EIs/discussion

By the way, the bug I reported many days ago about the "segmentation fault (core dump)" is still unsolved (see the google groups for details). I think this is a serious problem.

We are still working on this and will let you know when we find where the issue is and have the corresponding fix.

Thanks,

Sean

[Fengchao]

Hi Sean,

I use pwiz to parse the mzXML file which is converted from Water's raw data by ProteoWizard. I use the "Use zlib compression" during conversion. I think this bug is not a big deal because I could unselect "Use zlib compression" so that the converted mzXML file would be read correctly by crux.

As to the segmentation fault problem, it is serious because the software would crush if the data is large. I noticed that small precursor window will prevent this problem. Probably, it's out of memory or allocation error.  

Thank you,

Fengchao

Sean McIlwain

Tuesday, February 05, 2013 11:34 PM