Chem. J. Chinese Universities ›› 2002, Vol. 23 ›› Issue (4): 570.

• Articles • Previous Articles     Next Articles

Temperature-constrained Cascade Correlation Network andIts Application to Bacteria Identification

ZHANG Zhuo-Yong1, Aaron Urbas2, Peter de B Harrington2, Kent J. Voorhees3, Jon Rees3   

  1. 1. Department of Chemistry, Capital Normal University, Beijing 100037, China;
    2. Department of Chemistry and Biochemistry, Ohio University, Athens, Ohio 45701-2979, USA;
    3. Department of Geochemistry, Colorado School of Mines, Golden, CO 80401, USA
  • Received:2001-01-05 Online:2002-04-24 Published:2002-04-24

Abstract: The theoretical basis of temperature constrained cascade correlation network(TCCCN) was in-troduced briefly.The TCCCNmodels were used to identify various bacteria from their matrix-assistedlaser desorption/ionization time-of-flight mass spectrometric (MALDI-TOF-MS) data.The identificationof eight different bacteria and the bacteria identification based on different culture time have been dis-cussed.The results show that the TCCCNcould produce robust and feature-oriented models for bacteria i-dentification based on MALDI-TOF-MS.However, mis-identification may occur for some bacteria sam-ples due to variation in cell biology under environmental conditions.

Key words: Bacteria, MALDI, Time-of-flight mass spectrometry, TCCCN, Artificial neural network

CLC Number: 

TrendMD: