Neural network-based method for peptide identification in proteomics

Authors:
Lech Raczynski;Tymon Rubel;Krzysztof Zaremba
Affiliations:
Institute of Radioelectronics, Warsaw University of Technology, Warsaw, Poland;Institute of Radioelectronics, Warsaw University of Technology, Warsaw, Poland;Institute of Radioelectronics, Warsaw University of Technology, Warsaw, Poland
Venue:
ITIB'12 Proceedings of the Third international conference on Information Technologies in Biomedicine
Year:
2012

Citing 3
Cited 0

Feedforward Neural Network Methodology

Feedforward Neural Network Methodology
TANDEM: matching proteins with tandem mass spectra

Bioinformatics
Automatic Quality Assessment of Peptide Tandem Mass Spectra

Bioinformatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Protein identification in biological samples is one of the main objectives of proteomics. In proteomic experiments proteins are first digested into short peptides, which are next analyzed using tandem mass spectrometry and identified by database search algorithms. In this study a novel neural network-based method for peptide identification is proposed. The presented method improves the identification efficiency by the incorporation of additoinal peptide-specific features and scores from multiple database search algorithms. Moreover, the method for filtering out low quality mass spectra prior to database search in order to reduce the overall computational time of the identification process is presented.