<b><i>SuperHirn</i></b> – a novel tool for high resolution LC‐MS‐based peptide/protein profiling

Lukas Mueller, Oliver Rinner, Alexander Schmidt, Simon Letarte, Bernd Bodenmiller, Mi‐Youn Brusniak, Olga Vitek, Ruedi Aebersold, Markus Müller

PROTEOMICS · 2007 · 312 citations · 28 references

Concepts

TL;DR

Label‑free quantification of high‑resolution LC‑MS data has emerged as a promising technology for proteome analysis, requiring computational methods to accurately extract peptide signals and track them across samples. We present SuperHirn, an open‑source software tool comprising modules to process LC‑MS data acquired on a high‑resolution mass spectrometer. SuperHirn includes feature extraction and quantification, LC‑MS similarity analysis, alignment of multiple datasets, intensity normalization, and tools for clustering and classification of feature profiles. In a benchmark MS1‑based profiling of complex protein mixtures, SuperHirn automatically detected profiling trends and correctly associated proteins with their theoretical dilution profiles.

Abstract

Abstract Label‐free quantification of high mass resolution LC‐MS data has emerged as a promising technology for proteome analysis. Computational methods are required for the accurate extraction of peptide signals from LC‐MS data and the tracking of these features across the measurements of different samples. We present here an open source software tool, SuperHirn , that comprises a set of modules to process LC‐MS data acquired on a high resolution mass spectrometer. The program includes newly developed functionalities to analyze LC‐MS data such as feature extraction and quantification, LC‐MS similarity analysis, LC‐MS alignment of multiple datasets, and intensity normalization. These program routines extract profiles of measured features and comprise tools for clustering and classification analysis of the profiles. SuperHirn was applied in an MS1‐based profiling approach to a benchmark LC‐MS dataset of complex protein mixtures with defined concentration changes. We show that the program automatically detects profiling trends in an unsupervised manner and is able to associate proteins to their correct theoretical dilution profile.

References

28