Re: SINQ & normalized spectral index tools for label-free quantitation

[Greg Cary]

Micheleen,

That would be great!  Let me know if you get approval.

As I recall, the Normalized Spectral Index calculation (as outlined in the Nature Biotech paper) depends on Sequest-searched data (i.e. the DTAselect-filter.txt).  Does your python script make the same requirement? Or does it extract the MS2 intensities from other files?  I guess I can always re-search my data with Sequest...

Thanks so much!

Greg

On Friday, March 15, 2013 1:52:01 PM UTC-7, mha...@systemsbiology.org wrote:

Hi Greg,

We have implemented the Normalized Spectral Index (SIn) as a python script at ISB.  It is not available on regis yet as a production command line application, but the plan is to offer it in the future.  If you would like the code I could send it to you via DropBox as long as I get approval to do so.  Would you be interested in this?

Best,

Micheleen Harris

On Thursday, March 14, 2013 11:38:33 AM UTC-7, Greg Cary wrote:

Hi all,

I've been reading a lot lately about a variety of label-free quantitation techniques.  Unfortunately our data is poorly suited to MS1 peak extraction/integration (a la maxquant, skyline MS1 filtering etc.) and so we're very keen to implement some form of spectral counting (MS2 level) to compare complex co-purification from two different conditions.  It seems that in the world of spectral counting, recent movements have been toward some form of normalization that depends on extracting information from MS2 fragment ion intensities (e.g. SINQ, SIn,NSI/NSMT).  It appears that the CPFP at UT Southwestern has included SINQ quantitation as part of their dist via a perl module.  I'm wondering whether there are any tools/scripts available on regis that would allow for the extraction of MS2 fragment ion intensity values or if this was in consideration/development at all?

Thanks for any thoughts and comments on this approach!

Greg C. 

[Micheleen Harris]

Hi Greg,

You may search with any engine you wish (I've tested with X!Tandem,
Myrimatch, Comet, which is based on Sequest, and OMSSA). The NSI
application I've written needs the following(there are also a few
other parameters for input and output such as the directory of the
input files):

1) either mzML or mzXML files
2) an IProphet pep.xml (combining all your individual pep.xml files
from the search(s)) - The IProphet tool is available with the TPP.
3) the fasta file you used for the search
4) An FDR cutoff for the pep.xml file

Go ahead and do a search and run IProphet on the pep.xml files. I'll
let you know when the code is available.

Best,
Micheleen

> --
> You received this message because you are subscribed to a topic in the
> Google Groups "spctools-discuss" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/spctools-discuss/KbPZ6Ad4XLs/unsubscribe?hl=en.
> To unsubscribe from this group and all its topics, send an email to
> spctools-discu...@googlegroups.com.
> To post to this group, send email to spctools...@googlegroups.com.
> Visit this group at http://groups.google.com/group/spctools-discuss?hl=en.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

[Greg Cary]

Great Micheleen! I have all of the requisite data you mention as I have processed and analyzed my data using standard TPP protocols.

I'm looking forward to testing out this normalization on our data.

Greg

[Greg Cary]

Hi Dave,

Thanks for following up.  I had tried to use the CPFP demo server for a test with some of our data.  Unfortunately I never made it past the search steps in the workflow.  I just logged in and the two samples I'm trying to compare still report the status "SEARCHES QUEUED/RUNNING" and so I have not yet been able to attempt SINQ normalization.

If you let me know when things get sorted on your end, I'd love to try SINQ normalization as a comparison with our other approaches.

Thanks!

Greg

On Monday, April 1, 2013 2:27:27 PM UTC-7, Dave Trudgian wrote:

Hi Greg,

Just a quick note that if you happen to have tried SINQ on CPFP and have had any problems with the demo server I apologize. I had been away for a while and not really monitoring it, so may not have worked for you. Am going to spend some time on CPFP soon.

If you happen to try it on CPFP and have any problems then let me know.

DT

[Dave Trudgian]

Greg,

Sorry about that - I'll have it fixed tomorrow and contact you off list. The CPFP daemons are/were quite flaky, hence they break if any of various dependencies have any issue / the OS decides to upgrade and restart them etc. Am working on this, with help from others.

Will get back to you off list as this isn't a TPP matter itself.

DT