Multiple Search Engine Searches

Combining Multiple MS/MS Searches For Increased Peptide Identification

SEQUEST matches MS/MS spectra and peptides. So does Mascot. So does X!Tandem. Each of these search engines only matches fraction, usually 10, 20 or 30 percent, of the spectra. Typically all other spectra are discarded. One way that we can dig deeper to find information in these discarded spectra is to combine SEQUEST's matches with X!Tandem's. Or Mascot's with X!Tandem. Or SEQUEST with Mascot. Two searches dig up more information than one.


No Spectrum Left Behind

A video of the presentation that Brian Searle did at the NIH is a introduction to using multiple search engines to increase confidence in protein identifications.

In the accompanying slide show, Brian uses mass spectra from both QTOF and ion trap instruments from three different types of protein samples to show that searching with multiple search engines identifies significantly more spectra than using SEQUEST alone.

He also briefly explains how the program he created, Scaffold, uses Bayesian statistics to do this combination in an optimal way.

By taking advantage of the complementary strengths of different search engines, Scaffold identifies more of the spectra from a sample.

The result is that Scaffold identifies more proteins in a sample and more spectra for each protein.

In addition Scaffold focuses attention on those spectra which don't contribute to protein identifications that are most promising for further study.

Video of lecture at NIH

Improving Sensitivity by Combining Results from Multiple Search Methodologies

by Brian Searle

pdf format (0.5 MB)

PowerPoint format (3.5 MB)

29 slides

HUPO 2005 Poster

7th International Symposium on MS in Health/Life Sciences Poster

© Copyright 2009 Proteome Software, Inc.

1340 SW Bertha Blvd., Suite 10, Portland, OR 97219

1-800-944-6027    •