论文信息 - A genome wide RNAI screen by time lapse microscopy in order to identify mitotic genes — computational aspects and challenges

A genome wide RNAI screen by time lapse microscopy in order to identify mitotic genes — computational aspects and challenges

The MitoCheck project aims at identifying and characterizing the function of genes involved in mitosis in live human cells. The genome wide RNA interference screen developed for this purpose is based on automatic time-lapse microscopy to analyze the phenotypes after knocking down each human gene individually in cultured cells whose chromosomes are fluorescently labeled. Such a screen produces large amounts of digital image data (~ 200.000 video sequences, i.e. over 18 millions of images) which can no longer be handled and interpreted manually. We have developed an image processing method consisting of segmentation, feature extraction and automatic classification, which assigns to each nucleus in each image one out of several predefined morphological classes. Using the relative cell counts in each of these classes, measured over time for each experiment, we derive a phenotypic fingerprint for each gene that allows clustering of genes by functional similarity. This paper will give an overview over the computational aspects of this screen. The complete quality controlled data set and phenotypic measurements will be available after publication on http://www.mitocheck.org/.

[1] H. Erfle,et al. Reverse transfection on cell arrays for high content screening microscopy , 2007, Nature Protocols.

[2] Robert F. Murphy,et al. Robust Numerical Features for Description and Classification of Subcellular Location Patterns in Fluorescence Microscope Images , 2003, J. VLSI Signal Process..

[3] J. Ellenberg,et al. High-throughput fluorescence microscopy for systems biology , 2006, Nature Reviews Molecular Cell Biology.

[4] Meng Wang,et al. Novel cell segmentation and online SVM for cell cycle phase identification in automated microscopy , 2008, Bioinform..

[5] H. Erfle,et al. High-throughput RNAi screening by time-lapse imaging of live human cells , 2006, Nature Methods.

[6] Richard J. Prokop,et al. A survey of moment-based techniques for unoccluded object representation and recognition , 1992, CVGIP Graph. Model. Image Process..

[7] C. Conrad,et al. Automatic identification of subcellular phenotypes on human cell arrays. , 2004, Genome research.

[8] Robert M. Haralick,et al. Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[9] Paul T. Jackway,et al. Statistical geometric features-extensions for cytological texture analysis , 1996, Proceedings of 13th International Conference on Pattern Recognition.