Endogenous peptides / peptidomics

Search algorithms, post-searching processing, quantitation software, etc. Share and discuss software here.
m/z
Angiotensin Member
Angiotensin Member
Posts: 47
Joined: Thu Oct 17, 2013 5:06 am

Endogenous peptides / peptidomics

Postby m/z » Wed Feb 19, 2014 1:13 am

Hi!

Does anyone have experience with identification of endogenous peptides from tissue/cells?
As I understand, in a typical "peptidomics" workflow the peptides (~<10kDa) are simply separated from larger proteins and analyzed directly by MS (no digestion step is involved). The aim is often to identify smaller peptides that are generated from precursor peptides/proteins by regulated protease activity.
I imagine several challenges:
- it can be difficult to determine what is a biologically relevant peptide and what is simply the result of sample handling?
- a conventional proteomics software may not be the best option (since it is designed to identify proteins - not peptides) and wouldn't it be required to do a lot of critical and "manual" interpretation of spectra?
- some endogenous peptides are only three amino-acids long and and are probably easily missed?

Does anyone know of a nice (and free!) software that can help with peptidomics?
Any other comments regarding the analysis of endogenous peptides would also be greatly appreciated.
Thanks!

User avatar
Doug
E. Coli Lysate Member
E. Coli Lysate Member
Posts: 307
Joined: Sun Jun 26, 2011 7:20 pm

Postby Doug » Wed Feb 19, 2014 8:17 am

A coworker of mine worked on this awhile back. There are issues with ionization and fragmentation. Tryptic peptides are relatively short and contain basic moieties at both terminii making them well suited for LC-MS/MS. In order to make your sample amenable to analysis you may want to consider some type of chemical modification to introduce a positive charge (particularly at the C-terminal). You may also want to consider multiple fragmentation methods. ETD might be helpful here. Good luck.
"If we knew what we were doing it wouldn't be research." -AE

m/z
Angiotensin Member
Angiotensin Member
Posts: 47
Joined: Thu Oct 17, 2013 5:06 am

Postby m/z » Thu Feb 20, 2014 12:30 am

Thanks, Doug. Chemical modification sounds like a very good idea.

My other main concern is how to interpret the spectra - I am basically only familiar with normal proteomics software (mascot search, etc). For example, in Mascot you can select "none" for enzyme (instead of trypsin), but I wondering whether it is a bit naive to think that endogenous peptides can be found in this way? Well, I have some thinking to do :-)

Infinity
E. Coli Lysate Member
E. Coli Lysate Member
Posts: 107
Joined: Wed Dec 21, 2011 8:22 pm

Postby Infinity » Thu Feb 20, 2014 10:15 am

Well, I'm afraid it is the only one way to use unspecific cleavage. The problem here is that your search space will increase dramatically which will affect calculation time and also lead to underestimation of FDR.

Craig
E. Coli Lysate Member
E. Coli Lysate Member
Posts: 220
Joined: Sun Jun 26, 2011 6:49 pm

Postby Craig » Thu Feb 20, 2014 3:45 pm

Infinity wrote:The problem here is that your search space will increase dramatically which will affect calculation time and also lead to underestimation of FDR.


I don't think this will lead to underestimation of FDR. Your results will certainly be worse than if you could use enzyme specificity, but equally valid. Also keep in mind with such a large search space there is a higher-than-normal frequency of identical sequences between the target and decoy databases, so you might have to deal with that. But overall people have showed this strategy is feasible. High-resolution MS/MS would certainly help.

m/z
Angiotensin Member
Angiotensin Member
Posts: 47
Joined: Thu Oct 17, 2013 5:06 am

Postby m/z » Fri Feb 21, 2014 1:06 am

Thanks for the input.
One more question: What about identification of "micro-petides" of, say, only three residues. The smaller a peptide is the lower search score it will give right? So smaller peptides are more difficult to find. Or... ?

achimt
Phosphoserine Member
Phosphoserine Member
Posts: 23
Joined: Sun May 20, 2012 1:15 pm

Postby achimt » Fri Feb 21, 2014 1:41 am

The main problem with short peptides is that they could match many different proteins. ProteinProspector (http://prospector.ucsf.edu/prospector/cgi-bin/msform.cgi?form=mspattern) is one tool that you can use to estimate how often e.g. a particular tripeptide sequence will occur in a specific proteome. One example would be EHP, which occurs 16338 times in the human proteome.

m/z
Angiotensin Member
Angiotensin Member
Posts: 47
Joined: Thu Oct 17, 2013 5:06 am

Postby m/z » Fri Feb 21, 2014 2:00 am

Of cause. I didn't think of that. Thank you!

Craig
E. Coli Lysate Member
E. Coli Lysate Member
Posts: 220
Joined: Sun Jun 26, 2011 6:49 pm

Postby Craig » Fri Feb 21, 2014 8:41 am

m/z wrote:One more question: What about identification of "micro-petides" of, say, only three residues. The smaller a peptide is the lower search score it will give right? So smaller peptides are more difficult to find. Or... ?


Peptides this short are definitely very difficult to identify by mass spec. Since there are only 4 fragment ions to match for a 3-mer peptide, the chance of many or all of those being a spurious match is extremely high. For searches with no enzyme specificity I believe a minimum length of 5 or 6 is often used.

And as achimt said, peptides this short are usually unhelpful anyway because they lack protein specificity.

AAlpert
Angiotensin Member
Angiotensin Member
Posts: 34
Joined: Thu Jul 28, 2011 11:42 am

Postby AAlpert » Fri Feb 21, 2014 8:51 am

You might consider isolating the peptides in the 3-mer range by chromatography before attempting to identify them by MS. Our PolyHYDROXYETHYL A columns (100- or 200-Ã… pore), run in the SEC mode with 50 mM formic acid, will afford nice separations of peptides with 3-4, 5-6, or 7-8 residues. Alfred Goldberg's group at Harvard used them to identify the products of proteasome action, which are peptides in the 3-6 residue range. The columns are also routinely used to assess the distribution of small peptides in protein hydrolyzates used for food. Here's a link to my book chapter on the subject:
http://www.polylc.com/downloads/SEC_book_chapter_ver1.pdf

Andy Alpert
PolyLC Inc.

m/z
Angiotensin Member
Angiotensin Member
Posts: 47
Joined: Thu Oct 17, 2013 5:06 am

Postby m/z » Sat Feb 22, 2014 5:13 am

@Craig & AAlpert: Thanks for the advice and suggestions. I appreciate it.


Return to “Bioinformatics”

Who is online

Users browsing this forum: No registered users and 1 guest