A comparison of fuzzy clustering approaches for quantification of microarray gene expression

Yu Ping Wang, Maheswar Gunampally, Jie Chen, Douglas Bittel, Merlin G. Butler, Wei Wen Cai

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Despite the widespread application of microarray imaging for biomedical imaging research, barriers still exist regarding its reliability for clinical use. A critical major problem lies in accurate spot segmentation and the quantification of gene expression level (mRNA) from the microarray images. A variety of commercial and research freeware packages are available, but most cannot handle array spots with complex shapes such as donuts and scratches. Clustering approaches such as k-means and mixture models were introduced to overcome this difficulty, which use the hard labeling of each pixel. In this paper, we apply fuzzy clustering approaches for spot segmentation, which provides soft labeling of the pixel. We compare several fuzzy clustering approaches for microarray analysis and provide a comprehensive study of these approaches for spot segmentation. We show that possiblistic c-means clustering (PCM) provides the best performance in terms of stability criterion when testing on both a variety of simulated and real microarray images. In addition, we compared three statistical criteria in measuring gene expression levels and show that a new asymptotically unbiased statistic is able to quantify the gene expression level more accurately.

Original languageEnglish (US)
Pages (from-to)305-320
Number of pages16
JournalJournal of Signal Processing Systems
Volume50
Issue number3
DOIs
StatePublished - Mar 1 2008

Fingerprint

Fuzzy clustering
Fuzzy Clustering
Microarrays
Gene expression
Microarray
Quantification
Gene Expression
Segmentation
Labeling
Pixel
Clustering
Biomedical Imaging
Microarray Analysis
Pixels
K-means
Stability Criteria
Mixture Model
Imaging techniques
Messenger RNA
Statistic

Keywords

  • Fuzzy clustering
  • Image segmentation
  • Microarray
  • Microarray gridding
  • Segmentation

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Signal Processing
  • Information Systems
  • Modeling and Simulation
  • Hardware and Architecture

Cite this

A comparison of fuzzy clustering approaches for quantification of microarray gene expression. / Wang, Yu Ping; Gunampally, Maheswar; Chen, Jie; Bittel, Douglas; Butler, Merlin G.; Cai, Wei Wen.

In: Journal of Signal Processing Systems, Vol. 50, No. 3, 01.03.2008, p. 305-320.

Research output: Contribution to journalArticle

Wang, Yu Ping ; Gunampally, Maheswar ; Chen, Jie ; Bittel, Douglas ; Butler, Merlin G. ; Cai, Wei Wen. / A comparison of fuzzy clustering approaches for quantification of microarray gene expression. In: Journal of Signal Processing Systems. 2008 ; Vol. 50, No. 3. pp. 305-320.
@article{1986c595dc36469c87bcd0151eb55037,
title = "A comparison of fuzzy clustering approaches for quantification of microarray gene expression",
abstract = "Despite the widespread application of microarray imaging for biomedical imaging research, barriers still exist regarding its reliability for clinical use. A critical major problem lies in accurate spot segmentation and the quantification of gene expression level (mRNA) from the microarray images. A variety of commercial and research freeware packages are available, but most cannot handle array spots with complex shapes such as donuts and scratches. Clustering approaches such as k-means and mixture models were introduced to overcome this difficulty, which use the hard labeling of each pixel. In this paper, we apply fuzzy clustering approaches for spot segmentation, which provides soft labeling of the pixel. We compare several fuzzy clustering approaches for microarray analysis and provide a comprehensive study of these approaches for spot segmentation. We show that possiblistic c-means clustering (PCM) provides the best performance in terms of stability criterion when testing on both a variety of simulated and real microarray images. In addition, we compared three statistical criteria in measuring gene expression levels and show that a new asymptotically unbiased statistic is able to quantify the gene expression level more accurately.",
keywords = "Fuzzy clustering, Image segmentation, Microarray, Microarray gridding, Segmentation",
author = "Wang, {Yu Ping} and Maheswar Gunampally and Jie Chen and Douglas Bittel and Butler, {Merlin G.} and Cai, {Wei Wen}",
year = "2008",
month = "3",
day = "1",
doi = "10.1007/s11265-007-0123-0",
language = "English (US)",
volume = "50",
pages = "305--320",
journal = "Journal of Signal Processing Systems",
issn = "1939-8018",
publisher = "Springer New York",
number = "3",

}

TY - JOUR

T1 - A comparison of fuzzy clustering approaches for quantification of microarray gene expression

AU - Wang, Yu Ping

AU - Gunampally, Maheswar

AU - Chen, Jie

AU - Bittel, Douglas

AU - Butler, Merlin G.

AU - Cai, Wei Wen

PY - 2008/3/1

Y1 - 2008/3/1

N2 - Despite the widespread application of microarray imaging for biomedical imaging research, barriers still exist regarding its reliability for clinical use. A critical major problem lies in accurate spot segmentation and the quantification of gene expression level (mRNA) from the microarray images. A variety of commercial and research freeware packages are available, but most cannot handle array spots with complex shapes such as donuts and scratches. Clustering approaches such as k-means and mixture models were introduced to overcome this difficulty, which use the hard labeling of each pixel. In this paper, we apply fuzzy clustering approaches for spot segmentation, which provides soft labeling of the pixel. We compare several fuzzy clustering approaches for microarray analysis and provide a comprehensive study of these approaches for spot segmentation. We show that possiblistic c-means clustering (PCM) provides the best performance in terms of stability criterion when testing on both a variety of simulated and real microarray images. In addition, we compared three statistical criteria in measuring gene expression levels and show that a new asymptotically unbiased statistic is able to quantify the gene expression level more accurately.

AB - Despite the widespread application of microarray imaging for biomedical imaging research, barriers still exist regarding its reliability for clinical use. A critical major problem lies in accurate spot segmentation and the quantification of gene expression level (mRNA) from the microarray images. A variety of commercial and research freeware packages are available, but most cannot handle array spots with complex shapes such as donuts and scratches. Clustering approaches such as k-means and mixture models were introduced to overcome this difficulty, which use the hard labeling of each pixel. In this paper, we apply fuzzy clustering approaches for spot segmentation, which provides soft labeling of the pixel. We compare several fuzzy clustering approaches for microarray analysis and provide a comprehensive study of these approaches for spot segmentation. We show that possiblistic c-means clustering (PCM) provides the best performance in terms of stability criterion when testing on both a variety of simulated and real microarray images. In addition, we compared three statistical criteria in measuring gene expression levels and show that a new asymptotically unbiased statistic is able to quantify the gene expression level more accurately.

KW - Fuzzy clustering

KW - Image segmentation

KW - Microarray

KW - Microarray gridding

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=65349105333&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=65349105333&partnerID=8YFLogxK

U2 - 10.1007/s11265-007-0123-0

DO - 10.1007/s11265-007-0123-0

M3 - Article

AN - SCOPUS:65349105333

VL - 50

SP - 305

EP - 320

JO - Journal of Signal Processing Systems

JF - Journal of Signal Processing Systems

SN - 1939-8018

IS - 3

ER -