High-content Analysis in Monastrol Suppressor Screens

Summary

Objectives: High-content screening (HCS) via automated fluorescent microscopy is a powerful technology for the effective expression of cellular processes. However, HCS will generally produce tremendous image datasets, which leads to difficulties of handling and analyzing. We proposed an automatic classification approach for simultaneous feature extraction and cell phenotype recognition of monoaster and bipolar cells in HCS system.

Methods: The proposed approach was composed of image segmentation, feature extraction, and classification. The image segmentation was based on the Laplacian of Gaussian (LoG) edge detection method. For the reduction of noise effect on cellular images, we employed an adaptive threshold in microtubule channel. The principal component analysis was used in the feature selection process. The classification was performed with a back-propagation neural network (BPNN). Using the current approach, the cell phases were distinguished from three-channel acquisitions of cellular images and the numbers of bipolar and monoaster cells were automatically counted.

Results: The validity of this approach was examined by the application of screening the response of drug compounds in suppressing Monastrol. Our results indicate that the proposed algorithm could improve the recognition rates of monoaster and bipolar cells to 97.98% and 93.12%, respectively, compared with 97.02% and 86.96% obtained from the same samples by multi-phenotypic mitotic analysis (MMA).

Conclusions: We have shown that BPNN is a valuable tool to classify cell phenotype. To further improve the classification performance, more test data, more optimized feature selection approaches, and advanced classifier may be required and will be investigated in future works.

Keywords

High-content analysis - high-content screening - image analysis - back-propagation neural network - pattern recognition

• References

• 1 Cacace A. An ultra-HTS process for the identification of small molecule modulators of orphan G-protein-coupled receptors. Drug Discov Today 2003; 8: 785-792.
• 2 Dunkle R. Role of image informatics in accelerating drug discovery and development. Drug Discov World 2003; 5: 75-82.
• 3 Gaunitz F, Heise K. HTS compatible assay for anti-oxidative agents using primary cultured hepatocytes. Assay Drug Dev Technol 2003; 1: 469-477.
• 4 Klyushnichenko V. Protein crystallization: from HTS to kilo-gram-scale. Curr Opin Drug Discov Devel 2003; 6: 848-854.
• 5 Zhou XB, Wong STC. High content cellular imaging for drug development. IEEE Signal Processing Magazine 2006; 3: 1-5.
• 6 Murphy DB. Fundamentals of light microscopy and electronic imaging. New York: Wiley-Liss; 2001
• 7 Zhou XB, Cao XH, Perlman Z. A computerized cellular imaging system for high content analysis in Monastrol suppressor screens. J Biomed Inform 2005; 39: 115-125.
• 8 Zhou XB, Liu KL, Wong STC. Towards automated cellular image segmentation for RNAi genome-wide screening. Lecture Notes in Computer Science. Berlin: Springer; 2005; 3749: 885-892.
• 9 Chen XW, Zhou XB, Wong STC. Automated segmentation, classification, and tracking cell nuclei in time-lapse microscopy. IEEE Transactions on Biomedical Engineering 2006; 53: 762-766.
• 10 Chen XW, Wong STC. Knowledge-driven cell phase identification in time-lapse microscopy. IEEE Life Science Data Mining Workshop Brighton, UK: Nov. 2004
• 11 Chen XW. et al. An automated method for cell phase identification in high-throughput time-lapse screens. In: Wong STC, Li CS. (eds). Life Science Data Mining. World Scientific Publisher; Dec 2006
• 12 Zhou XB, Wong STC. Dynamic sub-celluar behavior study in high content imaging using a novel approach. Proc IEEE Conf Circuits and Systems. Kobe, Japan: May 2005
• 13 Zhou XB, Wong STC. Informatics Challenges of High-Throughput Microscopy. IEEE Signal Processing Magazine 2006; 23: 63-72.
• 14 Mayer TU. et al. Small molecule inhibitor of mitotic spindle bipolarity identified in a phenotype-based screen. Science 1999; 286: 971-974.
• 15 Kapoor TM, Mitchison TM. Eg5 is static in bipolar spindles relative to tubulin: evidence for a static spindle matrix. J Cell Biol 2001; 154: 971-974.
• 16 Gonzalez R, Woods R, Eddins S. Digital Image Processing Using MATLAB (2nd ed, Chinese simplified). Beijing: Publishing House of Electronics Industry, Prentice Hall International Inc.; 2005
• 17 Cohen I, Tian Q, Zhou XS, Huang TS. Feature selection using principal feature analysis. Proc 1st Int Conf on Image Processing 2002
• 18 Huang K, Murphy RF. Boosting accuracy of automated classification of fluorescence microscope images for location proteomics. BMC Bioinformatics 2004; 18: 78-84.
• 19 Boland MV, Murphy RF. A neural network classifier capable of recognizing the patterns of all major sub-cellular structures in fluorescence microscope images of HeLa cells. BMC Bioinformatics 2001; 17: 1213-1223.
• 20 Rubner J, Tavan P. A self-organizing network for principal-component analysis. Europhys Lett 1989; 10: 693-698.
• 21 Lindblad J. et al. Image Analysis for Automatic Segmentation of Cytoplasm and Classification of Rac1 Activation. Cytometry 2004; 57A: 22-33.
• 22 Theodoridis S, Koutroumbas K. Pattern recognition. San Diego Academic Press: 1999. pp 625-633.
• 23 Hornik KM, Stinchcombe M, White H. Multilayer feedforward networks are universal approximators. Neural Networks 1989; 2: 359-366.
• 24 Tao CY, Hoyt J, Feng Y. A support vector machine classifier for recognizing mitotic subphases using high-content screening data. Journal of Biomolecular Screening 2007; 12: 490-496.