A mass spectra-based compound-identification approach with a reduced reference library

Zhan Li Sun, Kin Man Lam, Jun Zhang

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

Abstract

In this paper, an effective and efficient compound identification approach is proposed based on the frequency feature of mass spectrum. A nonzero feature-retention strategy, and a correlation based-reference library reduction strategy, are designed in the proposed algorithm to reduce the computation burden. Further, a frequency feature based-composite similarity measure is adopted to decide the chemical abstracts service (CAS) registry numbers of mass spectral samples. Experimental results demonstrate the feasibility and efficiency of the proposed method.
Original languageEnglish
Title of host publicationIntelligent Computing Theories and Technology - 9th International Conference, ICIC 2013, Proceedings
Pages672-676
Number of pages5
DOIs
Publication statusPublished - 3 Sept 2013
Event9th International Conference on Intelligent Computing, ICIC 2013 - Nanning, China
Duration: 28 Jul 201331 Jul 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7996 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference9th International Conference on Intelligent Computing, ICIC 2013
Country/TerritoryChina
CityNanning
Period28/07/1331/07/13

Keywords

  • discrete Fourier transform
  • feature selection
  • similarity measure
  • Spectrum matching

ASJC Scopus subject areas

  • General Computer Science
  • Theoretical Computer Science

Fingerprint

Dive into the research topics of 'A mass spectra-based compound-identification approach with a reduced reference library'. Together they form a unique fingerprint.

Cite this