Spatiotemporal Gene Networks from ISH Images

As large-scale techniques for studying and measuring gene expressions have been developed, automatically inferring gene interaction networks from expression data has emerged as a popular technique to advance our understanding of cellular systems. Accurate prediction of gene interactions, especially in multicellular organisms such as Drosophila or humans, requires temporal and spatial analysis of gene expressions, which is not easily obtainable from microarray data. New image based techniques using in-situ hybridization(ISH) have recently been developed to allow large-scale spatial-temporal profiling of whole body mRNA expression. However, analysis of such data for discovering new gene interactions still remains an open challenge. This thesis studies the question of predicting gene interaction networks from ISH data. First, we present SpEX2, a computer vision pipeline to extract informative features from ISH data. Next, we present an algorithm, Gin-IM, for learning spatial gene interaction networks from embryonic ISH images at a single time step. Gin-IM combines multi-instance kernels with recent work in learning sparse undirected graphical models to predict interactions between genes. Next, we propose NP-MuScL (nonparanormal multi source learning) to estimate a gene interaction network that is consistent with multiple sources of data, having the same underlying relationships between the nodes. NP-MuScL uses the semiparametric Gaussian copula to model the distribution of the different data sources, with the different copulas sharing the same covariance matrix. Finally, we propose to extend NP-MuScL to also learn the importance or contribution of each data source in predicting the gene interaction network, using a few examples of known gene interactions. Weighted NP-MuScL uses the hinge loss to learn the contribution of each data source, with a KL-divergence penalty that encourages the weights to be similar to a prior for each data source. We also discuss reasonable priors for such a model. We propose to apply our algorithms on more than 100,000 Drosophila embryonic ISH images from the Berkeley Drosophila Genome Project. Each of the 6 time steps in Drosophila embryonic development is treated as a separate data source. With spatial gene interactions predicted via Gin-IM, and temporal predictions combined via weighted NP-MuScL, we will finally be able to predict spatiotemporal gene networks from these images.