A robust ratio estimator of gene expression via inverse-variance weighting for cDNA microarray images

Authors:
Shih-Huang Chan;Wan-Chi Chang
Affiliations:
Department of Statistics, National Cheng Kung University, No. 1, University Road, Tainan City 70101, Taiwan;Department of Statistics, National Cheng Kung University, No. 1, University Road, Tainan City 70101, Taiwan
Venue:
Computational Statistics & Data Analysis
Year:
2009

Citing 7
Cited 1

Seeded Region Growing

IEEE Transactions on Pattern Analysis and Machine Intelligence
Robust DNA microarray image analysis

Machine Vision and Applications
Spot shape modelling and data transformations for microarrays

Bioinformatics
Segmentation of cDNA Microarray Spots Using Markov Random Field Modeling

Bioinformatics
Segmentation of cDNA Microarray Spots Using Markov Random Field Modeling

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

Bioinformatics
Segmentation and intensity estimation of microarray images using a gamma-t mixture model

Bioinformatics

Editorial: Statistical genetics & statistical genomics: Where biology, epistemology, statistics, and computation collide

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.03

Visualization

Abstract

In microarray processing, the appearance of artifacts, donuts, and irregularly shaped spots is a problem. In current microarray analysis, most approaches stress the segmentation of pixel intensities rather than emphasizing ratio estimators. To avoid segmenting spot target areas and to minimize sensitivity to aberrant pixels, we propose a robust ratio estimator of gene expression via inverse-variance weighting. Moreover, a metric is proposed to evaluate the spot quality. Both the simulation and numerical examples explored reveal that the proposed algorithm is superior to existing approaches with respect to mean square error. The acceptance quality measure recommended confirms the validity of the proposed ratio estimator.