ABSTRACT
DNA microarray hybridisation is a popular high through-put technique in academic as well as industrial functional genomics research. In this paper we present a new approach to automatic grid segmentation of the raw fluorescence microarray images by Markov Random Field (MRF) techniques. The main objectives are applicability to various types of array designs and robustness to the typical problems encountered in microarray images, which are contaminations and weak signal.We briefly introduce microarray technology and give some background on MRFs. Our MRF model of microarray gridding is designed to integrate different application specific constraints and heuristic criteria into a robust and flexible segmentation algorithm. We show how to compute the model components efficiently and state our deterministic MRF energy minimization algorithm that was derived from the 'Highest Confidence First' algorithm by Chou et al. Since MRF segmentation may fail due to the properties of the data and the minimization algorithm, we use supplied or estimated print layouts to validate results.Finally we present results of tests on several series of microarray images from different sources, some of them test sets published with other microarray gridding software. Our MRF grid segmentation requires weaker assumptions about the array printing process than previously published methods and produces excellent results on many real datasets.An implementation of the described methods is available upon request from the authors.
- D. H. Ballard. Computer Vision. Prentice-Hall, Englewood Cliffs, 10. edition, 1982. Google ScholarDigital Library
- A. Blake and A. Zisserman. Visual Reconstruction. MIT Press, Cambridge, Mass., 1987. Google ScholarDigital Library
- D. Bozinov and J. Rahnenführer. Unsupervised technique for robust target separation and analysis of DNA microarray spots through adaptive pixel clustering. Bioinformatics, 18(5), 2002.Google Scholar
- C. S. Brown., P. Goodwin, and P. Sorger. Image metrics in the statistical analysis of DNA microarray data. PNAS, 98(16):8944--8949, 2001.Google ScholarCross Ref
- J. Buhler, T. Ideker, and D. Haynor. Dapple: Improved Techiques for Finding Spots on DNA Microarrays. Technical Report UWTR 2000-08-05, University of Washington, 2000.Google Scholar
- Y. Chen, E. R. Dougherty, and M. L. Bittner. Ratio-Based Decisions and the Quantitative Analysis of cDNA Microarray Images. J. Biomedical Optics, 2(4):364--374, 1997. Google ScholarDigital Library
- P. B. Chou, P. R. Cooper, M. J. Swain, C. M. Brown, and L. E. Wixson. Probabilistic Network Inference for Cooperative High and Low Level Vision. In R. Chellappa, editor, Markov random fields, pages 211--243. 1993.Google Scholar
- M. Eisen and P. Brown. DNA arrays for analysis of gene expression. Methods in Enzymology, 303, 1999.Google Scholar
- Gai, X., Lal, S., Xing, L., Brendel, V., and V. Walbot. Gene discovery using the maize genome database ZmDB. Nucleic Acids Research, (28):94--96, 2000.Google Scholar
- A. Jain, T. Tokuyasu, A. Snijders, R. Segraves, D. Albertson, and D. Pinkel. Fully Automatic Quantification of Microarray Image Data. Genome Res., 12(2):325--332, 2002.Google ScholarCross Ref
- M. Katzer, F. Kummert, and G. Sagerer. Automatische Auswertung von Mikroarraybildern. In Workshop Bildverarbeitung für die Medizin, Leipzig, 2002.Google ScholarCross Ref
- R. Kindermann and J. L. Snell. Markov random fields and their applications. Contemporary Mathematics 1. American Mathematical Society, Providence, RI, 1980.Google ScholarCross Ref
- S. Z. Li. Markov random field modeling in image analysis. Computer science workbench. Springer, Tokyo, 2. edition, 2001. Google ScholarDigital Library
- G. Sherlock, T. Hernandez-Boussard, and A. Kasarskis. The Stanford Microarray Database. Nucleic Acids Research, (29):152--155, 2001.Google Scholar
- M. Steinfath, W. Wruck, and H. Seidel. Automated image analysis for array hybridization experiments. Bioinformatics, 2001, Vol. 17, T. 7, S. 634--641, 2001.Google ScholarCross Ref
- Y. H. Yang, M. J. Buckley, S. Dudoit, and T. P. Speed. Comparison of Methods for Image Analysis on cDNA Microarray Data. Journal of Computational and Graphical Statistics, 11:108--136, 2002.Google ScholarCross Ref
Recommendations
Segmentation of cDNA Microarray Spots Using Markov Random Field Modeling
Motivation: Spot segmentation is a critical step in microarray gene expression data analysis. Therefore, the performance of segmentation may substantially affect the results of subsequent stages of the analysis, such as the detection of ...
The infinite hidden Markov random field model
Hidden Markov random field (HMRF) models are widely used for image segmentation, as they appear naturally in problems where a spatially constrained clustering scheme is asked for. A major limitation of HMRF models concerns the automatic selection of the ...
Comments