Intelligent Computing Theories and Technology

The relatively new technology of RNA interference (RNAi) can be used to suppress almost any gene on a genome wide scale. Fluorescence microscopy is an ever more advancing technology that allows for the visualization of cells in multidimensional fashion. The combination of these two techniques paired with automated image analysis is emerging as a powerful tool in system biology. It can be used to study the effects of gene knockdowns on cellular phenotypes thereby lightening shadow into the understanding of complex biological processes. In this paper we propose the use of high content screen (HCS) to derive a high quality functional gene network (FGN) for Drosophila Melanogaster. This approach is based on the characteristic patterns obtained from cell images under different gene knockdown conditions. We guarantee a high coverage of the resulting network by the further integration of a large set of heterogeneous genomic data. The integration of these diverse datasets is based on a linear support vector machine. The final network is analyzed and a signal transduction pathway for the mitogen-activated protein kinase (MAPK) pathway is extracted using an extended integer linear programming algorithm. We validate our results and demonstrate that the proposed method achieves full coverage of components deposited in KEGG for the MAPK pathway. Interestingly, we retrieved a set of additional candidate genes for this pathway (e.g. including sev, tsl, hb) that we suggest for future experimental validation.