MS Ana: A spectral library search engine optimized for high-accuracy fragment ion data

Aktivität: Gespräch oder VortragVortrag


Spectral library search uses spectrum-to-spectrum matching for the identification of peptides from fragment ion spectra. This approach is now experiencing growing interest in the mass spectrometry community thanks to the increasing number of available spectral libraries. Given a suitable library, using spectrum-tospectrum matching leads to higher sensitivity and faster processing times than database search1. However, the number of spectral library search engines that are readily available is still small. We present MS Ana: a spectral library search engine built to take advantage of libraries and experimental data with high-accuracy fragment ions. MS Ana uses an improved scoring function for spectrum-tospectrum matching in high-accuracy fragment ion data. The scoring uses several different statistical measures that focus on either peak mass or peak intensity and combines all of them to derive a scoring that makes best use of the high-accuracy data. We tested MS Ana performance on a variety of HeLA full cell digest HCD data using the NIST Human HCD spectral library. At FDR 1%, MS Ana identified on average 18.3% more unique peptides than database search with Sequest and 8.8% more unique peptides than to state-of-the-art spectral library search engine SpectraST. The prominent strategy for controlling FDR in proteomics experiments is the target-decoy approach that carries some issues for spectral library search since decoy library generation is not trivial. MS Ana allows for the generation of new decoy libraries using one of several different algorithms. Decoys can be quickly created for any spectral library independent of library structure or missing fragment annotations. MS Ana is available as a third-party node for the Thermo Fisher Scientific Proteome Discoverer and can be downloaded free-of-charge from Using the Proteome Discoverer software, setting up a search with MS Ana takes only minutes and allows for easy integration with additional analysis tools and existing workflows
Zeitraum16 Jän 2019
Ereignistitel2019 EuBIC Winter School on proteomics bioinformatics, Zakopane
OrtZakopane, Polen