Comparison of pattern detection methods in microarray time series of the segmentation clock

Mary Lee Dequéant, Sebastian Ahnert, Herbert Edelsbrunner, Thomas M A Fink, Earl F. Glynn, Gaye Hattem, Andrzej Kudlicki, Yuriy Mileyko, Jason Morton, Arcady R. Mushegian, Lior Pachter, Malgorzata Rowicka-Kudlicka, Anne Shiu, Bernd Sturmfels, Olivier Pourquié

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

While genome-wide gene expression data are generated at an increasing rate, the repertoire of approaches for pattern discovery in these data is still limited. Identifying subtle patterns of interest in large amounts of data (tens of thousands of profiles) associated with a certain level of noise remains a challenge. A microarray time series was recently generated to study the transcriptional program of the mouse segmentation clock, a biological oscillator associated with the periodic formation of the segments of the body axis. A method related to Fourier analysis, the Lomb-Scargle periodogram, was used to detect periodic profiles in the dataset, leading to the identification of a novel set of cyclic genes associated with the segmentation clock. Here, we applied to the same microarray time series dataset four distinct mathematical methods to identify significant patterns in gene expression profiles. These methods are called: Phase consistency, Address reduction, Cyclohedron test and Stable persistence, and are based on different conceptual frameworks that are either hypothesis- or data-driven. Some of the methods, unlike Fourier transforms, are not dependent on the assumption of periodicity of the pattern of interest. Remarkably, these methods identified blindly the expression profiles of known cyclic genes as the most significant patterns in the dataset. Many candidate genes predicted by more than one approach appeared to be true positive cyclic genes and will be of particular interest for future research. In addition, these methods predicted novel candidate cyclic genes that were consistent with previous biological knowledge and experimental validation in mouse embryos. Our results demonstrate the utility of these novel pattern detection strategies, notably for detection of periodic profiles, and suggest that combining several distinct mathematical approaches to analyze microarray datasets is a valuable strategy for identifying genes that exhibit novel, interesting transcriptional patterns.

Original languageEnglish (US)
Article numbere2856
JournalPLoS One
Volume3
Issue number8
DOIs
StatePublished - Aug 6 2008
Externally publishedYes

Fingerprint

Microarrays
Clocks
Time series
time series analysis
Genes
genes
Fourier Analysis
Gene expression
methodology
Biological Clocks
biological clocks
gene expression
mice
Periodicity
Microarray Analysis
Fourier analysis
Transcriptome
periodicity
Noise
Embryonic Structures

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Medicine(all)

Cite this

Dequéant, M. L., Ahnert, S., Edelsbrunner, H., Fink, T. M. A., Glynn, E. F., Hattem, G., ... Pourquié, O. (2008). Comparison of pattern detection methods in microarray time series of the segmentation clock. PLoS One, 3(8), [e2856]. https://doi.org/10.1371/journal.pone.0002856

Comparison of pattern detection methods in microarray time series of the segmentation clock. / Dequéant, Mary Lee; Ahnert, Sebastian; Edelsbrunner, Herbert; Fink, Thomas M A; Glynn, Earl F.; Hattem, Gaye; Kudlicki, Andrzej; Mileyko, Yuriy; Morton, Jason; Mushegian, Arcady R.; Pachter, Lior; Rowicka-Kudlicka, Malgorzata; Shiu, Anne; Sturmfels, Bernd; Pourquié, Olivier.

In: PLoS One, Vol. 3, No. 8, e2856, 06.08.2008.

Research output: Contribution to journalArticle

Dequéant, ML, Ahnert, S, Edelsbrunner, H, Fink, TMA, Glynn, EF, Hattem, G, Kudlicki, A, Mileyko, Y, Morton, J, Mushegian, AR, Pachter, L, Rowicka-Kudlicka, M, Shiu, A, Sturmfels, B & Pourquié, O 2008, 'Comparison of pattern detection methods in microarray time series of the segmentation clock', PLoS One, vol. 3, no. 8, e2856. https://doi.org/10.1371/journal.pone.0002856
Dequéant ML, Ahnert S, Edelsbrunner H, Fink TMA, Glynn EF, Hattem G et al. Comparison of pattern detection methods in microarray time series of the segmentation clock. PLoS One. 2008 Aug 6;3(8). e2856. https://doi.org/10.1371/journal.pone.0002856
Dequéant, Mary Lee ; Ahnert, Sebastian ; Edelsbrunner, Herbert ; Fink, Thomas M A ; Glynn, Earl F. ; Hattem, Gaye ; Kudlicki, Andrzej ; Mileyko, Yuriy ; Morton, Jason ; Mushegian, Arcady R. ; Pachter, Lior ; Rowicka-Kudlicka, Malgorzata ; Shiu, Anne ; Sturmfels, Bernd ; Pourquié, Olivier. / Comparison of pattern detection methods in microarray time series of the segmentation clock. In: PLoS One. 2008 ; Vol. 3, No. 8.
@article{1fc937e59320412592aa486434606c92,
title = "Comparison of pattern detection methods in microarray time series of the segmentation clock",
abstract = "While genome-wide gene expression data are generated at an increasing rate, the repertoire of approaches for pattern discovery in these data is still limited. Identifying subtle patterns of interest in large amounts of data (tens of thousands of profiles) associated with a certain level of noise remains a challenge. A microarray time series was recently generated to study the transcriptional program of the mouse segmentation clock, a biological oscillator associated with the periodic formation of the segments of the body axis. A method related to Fourier analysis, the Lomb-Scargle periodogram, was used to detect periodic profiles in the dataset, leading to the identification of a novel set of cyclic genes associated with the segmentation clock. Here, we applied to the same microarray time series dataset four distinct mathematical methods to identify significant patterns in gene expression profiles. These methods are called: Phase consistency, Address reduction, Cyclohedron test and Stable persistence, and are based on different conceptual frameworks that are either hypothesis- or data-driven. Some of the methods, unlike Fourier transforms, are not dependent on the assumption of periodicity of the pattern of interest. Remarkably, these methods identified blindly the expression profiles of known cyclic genes as the most significant patterns in the dataset. Many candidate genes predicted by more than one approach appeared to be true positive cyclic genes and will be of particular interest for future research. In addition, these methods predicted novel candidate cyclic genes that were consistent with previous biological knowledge and experimental validation in mouse embryos. Our results demonstrate the utility of these novel pattern detection strategies, notably for detection of periodic profiles, and suggest that combining several distinct mathematical approaches to analyze microarray datasets is a valuable strategy for identifying genes that exhibit novel, interesting transcriptional patterns.",
author = "Dequ{\'e}ant, {Mary Lee} and Sebastian Ahnert and Herbert Edelsbrunner and Fink, {Thomas M A} and Glynn, {Earl F.} and Gaye Hattem and Andrzej Kudlicki and Yuriy Mileyko and Jason Morton and Mushegian, {Arcady R.} and Lior Pachter and Malgorzata Rowicka-Kudlicka and Anne Shiu and Bernd Sturmfels and Olivier Pourqui{\'e}",
year = "2008",
month = "8",
day = "6",
doi = "10.1371/journal.pone.0002856",
language = "English (US)",
volume = "3",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "8",

}

TY - JOUR

T1 - Comparison of pattern detection methods in microarray time series of the segmentation clock

AU - Dequéant, Mary Lee

AU - Ahnert, Sebastian

AU - Edelsbrunner, Herbert

AU - Fink, Thomas M A

AU - Glynn, Earl F.

AU - Hattem, Gaye

AU - Kudlicki, Andrzej

AU - Mileyko, Yuriy

AU - Morton, Jason

AU - Mushegian, Arcady R.

AU - Pachter, Lior

AU - Rowicka-Kudlicka, Malgorzata

AU - Shiu, Anne

AU - Sturmfels, Bernd

AU - Pourquié, Olivier

PY - 2008/8/6

Y1 - 2008/8/6

N2 - While genome-wide gene expression data are generated at an increasing rate, the repertoire of approaches for pattern discovery in these data is still limited. Identifying subtle patterns of interest in large amounts of data (tens of thousands of profiles) associated with a certain level of noise remains a challenge. A microarray time series was recently generated to study the transcriptional program of the mouse segmentation clock, a biological oscillator associated with the periodic formation of the segments of the body axis. A method related to Fourier analysis, the Lomb-Scargle periodogram, was used to detect periodic profiles in the dataset, leading to the identification of a novel set of cyclic genes associated with the segmentation clock. Here, we applied to the same microarray time series dataset four distinct mathematical methods to identify significant patterns in gene expression profiles. These methods are called: Phase consistency, Address reduction, Cyclohedron test and Stable persistence, and are based on different conceptual frameworks that are either hypothesis- or data-driven. Some of the methods, unlike Fourier transforms, are not dependent on the assumption of periodicity of the pattern of interest. Remarkably, these methods identified blindly the expression profiles of known cyclic genes as the most significant patterns in the dataset. Many candidate genes predicted by more than one approach appeared to be true positive cyclic genes and will be of particular interest for future research. In addition, these methods predicted novel candidate cyclic genes that were consistent with previous biological knowledge and experimental validation in mouse embryos. Our results demonstrate the utility of these novel pattern detection strategies, notably for detection of periodic profiles, and suggest that combining several distinct mathematical approaches to analyze microarray datasets is a valuable strategy for identifying genes that exhibit novel, interesting transcriptional patterns.

AB - While genome-wide gene expression data are generated at an increasing rate, the repertoire of approaches for pattern discovery in these data is still limited. Identifying subtle patterns of interest in large amounts of data (tens of thousands of profiles) associated with a certain level of noise remains a challenge. A microarray time series was recently generated to study the transcriptional program of the mouse segmentation clock, a biological oscillator associated with the periodic formation of the segments of the body axis. A method related to Fourier analysis, the Lomb-Scargle periodogram, was used to detect periodic profiles in the dataset, leading to the identification of a novel set of cyclic genes associated with the segmentation clock. Here, we applied to the same microarray time series dataset four distinct mathematical methods to identify significant patterns in gene expression profiles. These methods are called: Phase consistency, Address reduction, Cyclohedron test and Stable persistence, and are based on different conceptual frameworks that are either hypothesis- or data-driven. Some of the methods, unlike Fourier transforms, are not dependent on the assumption of periodicity of the pattern of interest. Remarkably, these methods identified blindly the expression profiles of known cyclic genes as the most significant patterns in the dataset. Many candidate genes predicted by more than one approach appeared to be true positive cyclic genes and will be of particular interest for future research. In addition, these methods predicted novel candidate cyclic genes that were consistent with previous biological knowledge and experimental validation in mouse embryos. Our results demonstrate the utility of these novel pattern detection strategies, notably for detection of periodic profiles, and suggest that combining several distinct mathematical approaches to analyze microarray datasets is a valuable strategy for identifying genes that exhibit novel, interesting transcriptional patterns.

UR - http://www.scopus.com/inward/record.url?scp=51349152854&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51349152854&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0002856

DO - 10.1371/journal.pone.0002856

M3 - Article

VL - 3

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 8

M1 - e2856

ER -