Publication
Title
Comparison of the Mahalanobis distance and Pearson's statistic as measures of similarity of isotope patterns
Author
Abstract
To extract a genuine peptide signal from a mass spectrum, an observed series of peaks at a particular mass can be compared with the isotope distribution expected for a peptide of that mass. To decide whether the observed series of peaks is similar to the isotope distribution, a similarity measure is needed. In this short communication, we investigate whether the Mahalanobis distance could be an alternative measure for the commonly employed Pearson's chi(2) statistic. We evaluate the performance of the two measures by using a controlled MALDI-TOF experiment. The results indicate that Pearson's chi(2) statistic has better discriminatory performance than the Mahalanobis distance and is a more robust measure.
Language
English
Source (journal)
Journal of the American Society for Mass Spectrometry. - New York, N.Y.
Publication
New York, N.Y. : 2014
ISSN
1044-0305 [print]
1879-1123 [online]
DOI
10.1007/S13361-013-0773-Z
Volume/pages
25 :2 (2014) , p. 293-296
ISI
000330828700015
Full text (Publisher's DOI)
Full text (publisher's version - intranet only)
UAntwerpen
Faculty/Department
Research group
Publication type
Subject
Affiliation
Publications with a UAntwerp address
External links
Web of Science
Record
Identifier
Creation 25.03.2015
Last edited 09.10.2023
To cite this reference