More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies

Robert H. Baran, Changwon Jeon, David K. Han, Hanseok Ko

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Microbial DNA fragments are classified according to species using compositional features and "genomic signatures" the oldest of which is the dinucleotide relative abundance profile defined by Karlin et al. More informative features, including higher order signatures, have demonstrated greater species-specificity in comparison to the baseline established by the dinucleotide signature using "delta-distance" to assess dissimilarity; but lack of standard methods has precluded rigorous comparison. We describe a new method for classifier evaluation that reduces any number of pair-wise inter-genomic comparisons to a single performance measure. To illustrate the method, we compare delta-distance to quadratic and linear discriminants prescribed by elementary pattern recognition theory, and find that the quadratic form is significantly more powerful.

Original languageEnglish
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages605-608
Number of pages4
DOIs
Publication statusPublished - 2008 Sep 16
Event2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP - Las Vegas, NV, United States
Duration: 2008 Mar 312008 Apr 4

Other

Other2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
CountryUnited States
CityLas Vegas, NV
Period08/3/3108/4/4

Fingerprint

classifying
Pattern recognition
DNA
Classifiers
signatures
classifiers
pattern recognition
deoxyribonucleic acid
fragments
evaluation
profiles

Keywords

  • Biomedical signal processing
  • DNA
  • Error analysis
  • Pattern classification
  • Software performance

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Baran, R. H., Jeon, C., Han, D. K., & Ko, H. (2008). More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 605-608). [4517682] https://doi.org/10.1109/ICASSP.2008.4517682

More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies. / Baran, Robert H.; Jeon, Changwon; Han, David K.; Ko, Hanseok.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2008. p. 605-608 4517682.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Baran, RH, Jeon, C, Han, DK & Ko, H 2008, More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 4517682, pp. 605-608, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP, Las Vegas, NV, United States, 08/3/31. https://doi.org/10.1109/ICASSP.2008.4517682
Baran RH, Jeon C, Han DK, Ko H. More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2008. p. 605-608. 4517682 https://doi.org/10.1109/ICASSP.2008.4517682
Baran, Robert H. ; Jeon, Changwon ; Han, David K. ; Ko, Hanseok. / More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2008. pp. 605-608
@inproceedings{724fd56d8c04495ebc41e8c5816f726e,
title = "More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies",
abstract = "Microbial DNA fragments are classified according to species using compositional features and {"}genomic signatures{"} the oldest of which is the dinucleotide relative abundance profile defined by Karlin et al. More informative features, including higher order signatures, have demonstrated greater species-specificity in comparison to the baseline established by the dinucleotide signature using {"}delta-distance{"} to assess dissimilarity; but lack of standard methods has precluded rigorous comparison. We describe a new method for classifier evaluation that reduces any number of pair-wise inter-genomic comparisons to a single performance measure. To illustrate the method, we compare delta-distance to quadratic and linear discriminants prescribed by elementary pattern recognition theory, and find that the quadratic form is significantly more powerful.",
keywords = "Biomedical signal processing, DNA, Error analysis, Pattern classification, Software performance",
author = "Baran, {Robert H.} and Changwon Jeon and Han, {David K.} and Hanseok Ko",
year = "2008",
month = "9",
day = "16",
doi = "10.1109/ICASSP.2008.4517682",
language = "English",
isbn = "1424414849",
pages = "605--608",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - More powerful discriminants for classifying phylogenetic signals in dinucleotide frequencies

AU - Baran, Robert H.

AU - Jeon, Changwon

AU - Han, David K.

AU - Ko, Hanseok

PY - 2008/9/16

Y1 - 2008/9/16

N2 - Microbial DNA fragments are classified according to species using compositional features and "genomic signatures" the oldest of which is the dinucleotide relative abundance profile defined by Karlin et al. More informative features, including higher order signatures, have demonstrated greater species-specificity in comparison to the baseline established by the dinucleotide signature using "delta-distance" to assess dissimilarity; but lack of standard methods has precluded rigorous comparison. We describe a new method for classifier evaluation that reduces any number of pair-wise inter-genomic comparisons to a single performance measure. To illustrate the method, we compare delta-distance to quadratic and linear discriminants prescribed by elementary pattern recognition theory, and find that the quadratic form is significantly more powerful.

AB - Microbial DNA fragments are classified according to species using compositional features and "genomic signatures" the oldest of which is the dinucleotide relative abundance profile defined by Karlin et al. More informative features, including higher order signatures, have demonstrated greater species-specificity in comparison to the baseline established by the dinucleotide signature using "delta-distance" to assess dissimilarity; but lack of standard methods has precluded rigorous comparison. We describe a new method for classifier evaluation that reduces any number of pair-wise inter-genomic comparisons to a single performance measure. To illustrate the method, we compare delta-distance to quadratic and linear discriminants prescribed by elementary pattern recognition theory, and find that the quadratic form is significantly more powerful.

KW - Biomedical signal processing

KW - DNA

KW - Error analysis

KW - Pattern classification

KW - Software performance

UR - http://www.scopus.com/inward/record.url?scp=51449100134&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51449100134&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2008.4517682

DO - 10.1109/ICASSP.2008.4517682

M3 - Conference contribution

AN - SCOPUS:51449100134

SN - 1424414849

SN - 9781424414840

SP - 605

EP - 608

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -