Donuts, scratches and blanks: Robust model-based segmentation of microarray images

Qunhua Li, Chris Fraley, Roger E. Bumgarner, Ka Yee Yeung, Adrian E. Raftery

Research output: Contribution to journalArticle

79 Citations (Scopus)

Abstract

Motivation: Inner holes, artifacts and blank spots are common in microarray images, but current image analysis methods do not pay them enough attention. We propose a new robust model-based method for processing microarray images so as to estimate foreground and background intensities. The method starts with a very simple but effective automatic gridding method, and then proceeds in two steps. The first step applies model-based clustering to the distribution of pixel intensities, using the Bayesian Information Criterion (BIC) to choose the number of groups up to a maximum of three. The second step is spatial, finding the large spatially connected components in each cluster of pixels. The method thus combines the strengths of the histogram-based and spatial approaches. It deals effectively with inner holes in spots and with artifacts. It also provides a formal inferential basis for deciding when the spot is blank, namely when the BIC favors one group over two or three. Results: We apply our methods for gridding and segmentation to cDNA microarray images from an HIV infection experiment. In these experiments, our method had better stability across replicates than a fixed-circle segmentation method or the seeded region growing method in the SPOT software, without introducing noticeable bias when estimating the intensities of differentially expressed gene.

Original languageEnglish (US)
Pages (from-to)2875-2882
Number of pages8
JournalBioinformatics
Volume21
Issue number12
DOIs
StatePublished - Jun 15 2005

Fingerprint

Microarrays
Microarray
Segmentation
Model-based
Pixels
Image analysis
Bayesian Information Criterion
Complementary DNA
Genes
Experiments
Artifacts
Processing
Pixel
CDNA Microarray
Model-based Clustering
HIV Infection
Region Growing
Oligonucleotide Array Sequence Analysis
Connected Components
Image Analysis

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

Li, Qunhua ; Fraley, Chris ; Bumgarner, Roger E. ; Yeung, Ka Yee ; Raftery, Adrian E. / Donuts, scratches and blanks : Robust model-based segmentation of microarray images. In: Bioinformatics. 2005 ; Vol. 21, No. 12. pp. 2875-2882.
@article{c3c62c50ebec48a88800e861312532b8,
title = "Donuts, scratches and blanks: Robust model-based segmentation of microarray images",
abstract = "Motivation: Inner holes, artifacts and blank spots are common in microarray images, but current image analysis methods do not pay them enough attention. We propose a new robust model-based method for processing microarray images so as to estimate foreground and background intensities. The method starts with a very simple but effective automatic gridding method, and then proceeds in two steps. The first step applies model-based clustering to the distribution of pixel intensities, using the Bayesian Information Criterion (BIC) to choose the number of groups up to a maximum of three. The second step is spatial, finding the large spatially connected components in each cluster of pixels. The method thus combines the strengths of the histogram-based and spatial approaches. It deals effectively with inner holes in spots and with artifacts. It also provides a formal inferential basis for deciding when the spot is blank, namely when the BIC favors one group over two or three. Results: We apply our methods for gridding and segmentation to cDNA microarray images from an HIV infection experiment. In these experiments, our method had better stability across replicates than a fixed-circle segmentation method or the seeded region growing method in the SPOT software, without introducing noticeable bias when estimating the intensities of differentially expressed gene.",
author = "Qunhua Li and Chris Fraley and Bumgarner, {Roger E.} and Yeung, {Ka Yee} and Raftery, {Adrian E.}",
year = "2005",
month = "6",
day = "15",
doi = "10.1093/bioinformatics/bti447",
language = "English (US)",
volume = "21",
pages = "2875--2882",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "12",

}

Donuts, scratches and blanks : Robust model-based segmentation of microarray images. / Li, Qunhua; Fraley, Chris; Bumgarner, Roger E.; Yeung, Ka Yee; Raftery, Adrian E.

In: Bioinformatics, Vol. 21, No. 12, 15.06.2005, p. 2875-2882.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Donuts, scratches and blanks

T2 - Robust model-based segmentation of microarray images

AU - Li, Qunhua

AU - Fraley, Chris

AU - Bumgarner, Roger E.

AU - Yeung, Ka Yee

AU - Raftery, Adrian E.

PY - 2005/6/15

Y1 - 2005/6/15

N2 - Motivation: Inner holes, artifacts and blank spots are common in microarray images, but current image analysis methods do not pay them enough attention. We propose a new robust model-based method for processing microarray images so as to estimate foreground and background intensities. The method starts with a very simple but effective automatic gridding method, and then proceeds in two steps. The first step applies model-based clustering to the distribution of pixel intensities, using the Bayesian Information Criterion (BIC) to choose the number of groups up to a maximum of three. The second step is spatial, finding the large spatially connected components in each cluster of pixels. The method thus combines the strengths of the histogram-based and spatial approaches. It deals effectively with inner holes in spots and with artifacts. It also provides a formal inferential basis for deciding when the spot is blank, namely when the BIC favors one group over two or three. Results: We apply our methods for gridding and segmentation to cDNA microarray images from an HIV infection experiment. In these experiments, our method had better stability across replicates than a fixed-circle segmentation method or the seeded region growing method in the SPOT software, without introducing noticeable bias when estimating the intensities of differentially expressed gene.

AB - Motivation: Inner holes, artifacts and blank spots are common in microarray images, but current image analysis methods do not pay them enough attention. We propose a new robust model-based method for processing microarray images so as to estimate foreground and background intensities. The method starts with a very simple but effective automatic gridding method, and then proceeds in two steps. The first step applies model-based clustering to the distribution of pixel intensities, using the Bayesian Information Criterion (BIC) to choose the number of groups up to a maximum of three. The second step is spatial, finding the large spatially connected components in each cluster of pixels. The method thus combines the strengths of the histogram-based and spatial approaches. It deals effectively with inner holes in spots and with artifacts. It also provides a formal inferential basis for deciding when the spot is blank, namely when the BIC favors one group over two or three. Results: We apply our methods for gridding and segmentation to cDNA microarray images from an HIV infection experiment. In these experiments, our method had better stability across replicates than a fixed-circle segmentation method or the seeded region growing method in the SPOT software, without introducing noticeable bias when estimating the intensities of differentially expressed gene.

UR - http://www.scopus.com/inward/record.url?scp=20844437272&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=20844437272&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/bti447

DO - 10.1093/bioinformatics/bti447

M3 - Article

C2 - 15845656

AN - SCOPUS:20844437272

VL - 21

SP - 2875

EP - 2882

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 12

ER -