Regularized kernel discriminant analysis (RKDA) performs linear discriminant analysis in the feature space via the kernel trick. Its performance depends on the selection of kernels. In this paper, we consider the problem of multiple kernel learning (MKL) for RKDA, in which the optimal kernel matrix is obtained as a linear combination of pre-specified kernel matrices. We show that the kernel learning problem in RKDA can be formulated as convex programs. First, we show that this problem can be formulated as a semidefinite program (SDP). Based on the equivalence relationship between RKDA and least square problems in the binary-class case, we propose a convex quadratically constrained quadratic programming (QCQP) formulation for kernel learning in RKDA. A semi-infinite linear programming (SILP) formulation is derived to further improve the efficiency. We extend these formulations to the multi-class case based on a key result established in this paper. That is, the multi-class RKDA kernel learning problem can be decomposed into a set of binary-class kernel learning problems which are constrained to share a common kernel. Based on this decomposition property, SDP formulations are proposed for the multi-class case. Furthermore, it leads naturally to QCQP and SILP formulations. As the performance of RKDA depends on the regularization parameter, we show that this parameter can also be optimized in a joint framework with the kernel. Extensive experiments have been conducted and analyzed, and connections to other algorithms are discussed.

To compare spatial patterns of gene expression, one must analyze a large number of images as current methods are only able to measure a small number of genes at a time. Bringing images of corresponding tissues into alignment is a critical first step in making a meaningful comparative analysis of these spatial patterns. Significant image noise and variability in the shapes make it hard to pick a canonical shape model. In this paper, we address these problems by combining segmentation and unsupervised shape learning algorithms. We first segment images to acquire structures of interest, then jointly align the shapes of these acquired structures using an unsupervised nonparametric maximum likelihood algorithm along the lines of ‘congealing’ [12], while simultaneously learning the underlying shape model and associated transformations. The learned transformations are applied to corresponding images to bring them into alignment in one step. We demonstrate the results for images of various classes of Drosophila imaginal discs and discuss the methodology used for a quantitative analysis of spatial gene expression patterns. 1.

The Berkeley Drosophila Genome Project (BDGP) has produced a large number of gene expression patterns, many of which have been annotated textually with anatomical and developmental terms. These terms spatially correspond to local regions of the images; however, they are attached collectively to groups of images, such that it is unknown which term is assigned to which region of which image in the group. This poses a challenge to the development of the computational method to automate the textual description of expression patterns contained in each image. In this paper, we show that the underlying nature of this task matches well with a new machine learning framework, Multi-Instance Multi-Label learning (MIML). We propose a new MIML support vector machine to solve the problems that beset the annotation task. Empirical study shows that the proposed method outperforms the state-of-the-art Drosophila gene expression pattern annotation methods. 1

We present an application of image analysis techniques to automatically annotate biological images depicting gene expression patterns in developing embryos of fruit fly (Drosophila melanogaster), a model organism to study gene interaction. The aim is to determine the view (lateral versus dorsal/ventral [non-lateral]), orientation (anterior-left or anterior-right), and the developmental stage of the embryo. We employed contour curvature analysis, symmetry of the gene expression patterns, and shape differences at the anterior and posterior ends of the embryo, among others, for these purposes. An analysis of a pilot database of 3500 images indicates that view was correctly identified in 62%, orientation in 85%, and developmental stage in 73 % of the images. We observed that correct inferences had better separation in feature space than incorrect inferences. This means that, although these methods do not exhibit very high classification accuracy, they could be employed to identify images which need manual intervention, thereby reducing the target set for biologists. The novelty in this work is in the integration of well-established image analysis with the biological knowledge for annotating the embryos. Our examinations show that features that provide discrimination ability among different views, different orientations, and different developmental stages are often restricted to certain regions of the

and in the United Kingdom by Information Science Reference (an imprint of IGI Global)

High-throughput experiments have produced convicing evidence for an extensive contribution of diverse classes of RNAs in the expression of genetic information. Instead of a simple arrangement of mostly protein-coding genes, the human transcriptome features a complex arrangement of overlapping transcripts, many of which do not code for proteins at all, while others “sample ” exons from several different “genes”. The complexity of the transcriptome and the prevalence of noncoding transcripts forces us to reconsider both the concept of the “gene ” itself and our understanding of the mechanisms that regulate “gene expression”. 1.

Abstract. Carefully studied in-situ hybridization Gene expression patterns (GEP) can provide a first glance at possible relationships among genes. Automatic comparative analysis tools are an indispensable requirement to manage the constantly growing amount of such GEP images. We present here an automated processing pipeline for Segmenting, Classification, and Clustering large-scale sets of Drosophila melanogaster GEP images that facilitates automatic GEP analysis. 1