Combine Multiple Mass Spectral Similarity Measures for Compound Identification in GC-MS

In this study, total seven similarity measures were combined to improve the identification performance. To test the developed system, 28,234 mass spectra from the NIST replicate library were randomly split into the training set and test set. PSO algorithm was used to find the optimized weights of se...

Full description

Saved in:
Bibliographic Details
Published inIntelligent Computing in Bioinformatics pp. 255 - 261
Main Authors Liao, Li-Huan, Zhu, Yi-Fei, Cao, Li-Li, Zhang, Jun
Format Book Chapter
LanguageEnglish
Published Cham Springer International Publishing 2014
SeriesLecture Notes in Computer Science
Subjects
compound identification
mass spectra
similarity measures
Online AccessGet full text
ISBN9783319093291
3319093290
ISSN0302-9743
1611-3349
DOI10.1007/978-3-319-09330-7_31

Cover

More Information
Summary:In this study, total seven similarity measures were combined to improve the identification performance. To test the developed system, 28,234 mass spectra from the NIST replicate library were randomly split into the training set and test set. PSO algorithm was used to find the optimized weights of seven similarity measures based on the training set, and then the optimized weights were applied into the test set. Simulation study indicates that the combination of multiple similarity measures achieves a better performance than single best measure, with the identification accuracy improved by 2.2 % and 1.7% for training and test set respectively.
ISBN:9783319093291
3319093290
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-319-09330-7_31