xpress has an explicit 'label free' option that should give you virtually
the same precursor peak areas as you got using the option you used above.

The -M option you used was for N15 metabolic labeling where every amino
acid residue is modified.  Because every peptide has a modification, this
had the side effect of forcing xpress to quantify every peptide ID.  And by
using the "light_area" peak areas, this gives you the precursor peak areas.
 There is now a "-l" label free mode which generates a precursor peak area
for every peptide.  Your command line would look something like:

   xinteract -x20 -Opt -PREC -X-l-m0.05 -NResult.pep.xml
somefile.raw.pep.xml

Answers to your specific questions:

1.  The "-M" option will cause xpress to quantify every peptide; you would
not need to use the "-n" option too.  However, you can save a little
computation time (by not quantifying a non-existent heavy pair) by using
the "-l" label free option instead.

2.  The optimal "-m" mass tolerance option is some setting that you'll need
to evaluate yourself.  I would stick with 0.05 myself.  Another developer
has added support to represent the mass tolerance in terms of ppm using the
"-a" option.  However, this new option is currently in the repository trunk
code and not part of any TPP release.  You will need to compile
XPressPeptideParser yourself from the trunk code to gain access to this new
option.

3.  This setting depends on what you want; there's really no right/wrong
choice.  When this value is set to 6, it requires a chromatogram to have at
least 6 ms1 peaks in it for the peak area to be reported.  Otherwise xpress
will not export a precursor peak area for those chromatograms that have
less ms1 peaks than you choose in this setting.  Change this value to 1 if
you want xpress to report a peak area for everything, even if there's only
1 ms1 signal in the elution peak.

4.  No.  Different peptides from the same protein ionize differently and
this commonly gives a wide distribution of peptide peak areas from the same
protein.  So I don't even know what the right choice would be to take
disparate raw peptide peak areas to somehow represent a protein number.

For precursor peak area, xpress simply sums the MS1 precursor peak
intensities across some time range and reports this summed intensity.  I
don't know what Xcalibur does when it reports a peak area but I do know the
range if reported numbers are different.  If you're looking for alternative
tools to do label free analysis, I've been told that MassChroQ works well.
 I believe Skyline will do this for you too.  Good luck.

- Jimmy


On Fri, Sep 13, 2013 at 9:24 AM, <[email protected]> wrote:

> I am using xpress algorithms to get peak areas under curve (label free). I
> did some homework in this forum, and it turns out that it is doable by
> xpress according to this post:
>
>
> https://groups.google.com/forum/#!searchin/spctools-discuss/label$20free$20xpress/spctools-discuss/RQPvyz3hIa4/bAVLoHxjinwJ
>
> In the above post, it is suggested to run xinteract with -X-M option, and
> then grab only "light area" in the pepxml file. So I did it using the
> command:
>
> xinteract -x20 -Opt -PREC -X-M-m0.05 -NResult.pep.xml somefile.raw.pep.xml
> (tpp v4.5 rapture rev 2)
>
> Inside Result.pep.xml file, I found description of xpress algorithm:
>
> <xpressratio_summary version="2.1 (TPP v4.5 RAPTURE rev 2, Build
> 201307022022 (linux))" author="Jimmy Eng" same_scan_range="Y"
> labeled_residues="MLIGHT" xpress_light="0" massdiff="MLIGHT,1.00"
> masstol="0.05" min_num_chromatogram_points="6" min_num_isotope_peaks="0"/>
>
> My questions are:
>
>    1. Basically -M option defaults "MLIGHT" as labeled, not every amino
>    acid; Should I indicates all amino acid residues by using -n option?
>    2. My raw data has been searched by setting parent mass tolerance of
>    20ppm. I am wondering if it is reasonable to set mass tolerance to 0.05
>    dalton here by using -m option.
>    3. for: min_num_chromatogram_points="6", is it to leave it just like
>    this or change to some other values?
>    4. Is there any way to get protein level quantitation by use of peak
>    area of peptides?
>
>
> I have these questions since, when I manually check the peak areas by
> xcalibur, it turns out to be very different than what xpress gives to me.
> So I figured that something must be wrong with my setting. Can somebody
> help me out on these issues?
>
>
>  --
> 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