Automated annotation of Drosophila gene expression patterns using a controlled vocabulary

MOTIVATION: Regulation of gene expression in space and time directs its localization to a specific subset of cells during development. Systematic determination of the spatiotemporal dynamics of gene expression plays an important role in understanding the regulatory networks driving development. An atlas for the gene expression patterns of fruit fly Drosophila melanogaster has been created by whole-mount in situ hybridization, and it documents the dynamic changes of gene expression pattern during Drosophila embryogenesis. The spatial and temporal patterns of gene expression are integrated by anatomical terms from a controlled vocabulary linking together intermediate tissues developed from one another. Currently, the terms are assigned to patterns manually. However, the number of patterns generated by high-throughput in situ hybridization is rapidly increasing. It is, therefore, tempting to approach this problem by employing computational methods. RESULTS: In this article, we present a novel computational framework for annotating gene expression patterns using a controlled vocabulary. In the currently available high-throughput data, annotation terms are assigned to groups of patterns rather than to individual images. We propose to extract invariant features from images, and construct pyramid match kernels to measure the similarity between sets of patterns. To exploit the complementary information conveyed by different features and incorporate the correlation among patterns sharing common structures, we propose efficient convex formulations to integrate the kernels derived from various features. The proposed framework is evaluated by comparing its annotation with that of human curators, and promising performance in terms of F1 score has been reported.     

Automatic recognition and annotation of gene expression patterns of fly embryos

MOTIVATION: Gene expression patterns obtained by in situ mRNA hybridization provide important information about different genes during Drosophila embryogenesis. So far, annotations of these images are done by manually assigning a subset of anatomy ontology terms to an image. This time-consuming process depends heavily on the consistency of experts. RESULTS: We develop a system to automatically annotate a fruitfly's embryonic tissue in which a gene has expression. We formulate the task as an image pattern recognition problem. For a new fly embryo image, our system answers two questions: (1) Which stage range does an image belong to? (2) Which annotations should be assigned to an image? We propose to identify the wavelet embryo features by multi-resolution 2D wavelet discrete transform, followed by min-redundancy max-relevance feature selection, which yields optimal distinguishing features for an annotation. We then construct a series of parallel bi-class predictors to solve the multi-objective annotation problem since each image may correspond to multiple annotations. SUPPLEMENTARY INFORMATION: The complete annotation prediction results are available at: http://www.cs.niu.edu/~jzhou/papers/fruitfly and http://research.janelia.org/peng/proj/fly_embryo_annotation/. The datasets used in experiments will be available upon request to the correspondence author.     

Automatic recognition and annotation of gene expression patterns of fly embryos

Vol. 23, No.     

Clustering of spatial gene expression patterns in the mouse brain and comparison with classical neuroanatomy

Spatial gene expression profiles provide a novel means of exploring the structural organization of the brain. Computational analysis of these patterns is made possible by genome-scale mapping of the C57BL/6J mouse brain in the Allen Brain Atlas. Here we describe methodology used to explore the spatial structure of gene expression patterns across a set of 3041 genes chosen on the basis of consistency across experimental observations (N = 2). The analysis was performed on smoothed, co-registered 3D expression volumes for each gene obtained by aggregating cellular resolution image data. Following dimensionality and noise reduction, voxels were clustered according to similarity of expression across the gene set. We illustrate the resulting parcellations of the mouse brain for different numbers of clusters (K) and quantitatively compare these parcellations with a classically-defined anatomical reference atlas at different levels of granularity, revealing a high degree of correspondence. These observations suggest that spatial localization of gene expression offers substantial promise in connecting knowledge at the molecular level with higher-level information about brain organization.     

Spatial patterns of gene expression in preimplantation mouse embryos

The distribution of total polyadenylated RNA and mRNAs from the beta-actin, fibronectin, and cytokeratin Endo A genes was examined in preimplantation mouse embryos using in situ hybridization of riboprobes to RNA in sections of embryos. Polyadenylated RNA was found in the cytoplasm of all cells of blastocyst-stage embryos, whereas the specific mRNAs displayed three distinct patterns of expression: uniform throughout the embryo (beta-actin), enriched in the inner cell mass (fibronectin), and enriched in the trophectoderm (Endo A). In eight-cell embryos, the polyadenylated RNA was more concentrated in nuclei than in the cytoplasm (as noted previously), although this was not the case in blastocysts, nor was it true for the specific mRNAs that were examined. These experiments demonstrate that there is localized gene expression in the early mouse embryo, which correlates with the formation of the trophectoderm and the inner cell mass.     

Evaluation of 2-DE protein patterns from pre- and postnatal stages of the mouse brain

Brains of the mouse from three developmental stages, embryo day 16 (Ed16), postnatal stage one week (1W) and eight weeks (8W), were distributed to different laboratories for a collaborative proteome analysis (The Human Brain Proteome Project). As one of the laboratories involved in this project, we separated total protein extracts of the brains by large gel 2-DE. From the 2-DE protein patterns a section was evaluated for each of the three stages according to resolution, reproducibility and quantitative changes using an image analysis software. The evaluated pattern section was selected to allow comparisons of 2-DE patterns between different laboratories on the basis of optimum separation. Changes in protein expression were analysed within two phases of development: Stage Ed16 versus stage 1W and stage 1W versus stage 8W. Out of the 200 protein spots evaluated 5-6% showed quantitative changes in the range of > or = 30% between two stages. The relationship in the frequency of up- and down-regulated protein spots differed between the two investigated phases. Most of the protein spots which showed altered expression between two stages were identified by MS. High quality in protein separation and evaluation is demonstrated.     

Early gene expression profile in mouse brain after exposure to ionizing radiation

Acute changes in the gene expression profile in mouse brain after exposure to ionizing radiation were studied using microarray analysis. RNA was isolated at 0.25, 1, 5 and 24 h after exposure to 20 Gy and at 5 h after exposure of the whole brain of adult mice to 2 or 10 Gy. RNA was hybridized onto 15K cDNA microarrays, and data were analyzed using GeneSpring and Significant Analysis of Microarray. Radiation modulated the expression of 128, 334, 325 and 155 genes and ESTs at 0.25, 1, 5 and 24 h after 20 Gy and 60 and 168 at 5 h after 2 and 10 Gy, respectively. The expression profiles showed dose- and time-dependent changes in both expression levels and numbers of differentially modulated genes and ESTs. Seventy-eight genes were modulated at two or more times. Differentially modulated genes were associated with 12 different classes of molecular function and 24 different biological pathways and showed time- and dose-dependent changes. The change in expression of four genes (Jak3, Dffb, Nsep1 and Terf1) after irradiation was validated using quantitative real-time PCR. Up-regulation of Jak3 was observed in another mouse strain. In mouse brain, there was an increase of Jak3 immunoreactivity after irradiation. In conclusion, changes in the gene profile in the brain after irradiation are complex and are dependent on time and dose, and genes with diverse functions and pathways are modulated.     

A review of gene expression patterns in the malformed brain

Major advances in the identification of genes expressed in malformation-associated epileptic disorders have been made. Some of these changes reflect the complex gene interactions necessary for proper neurodevelopment, whereas others suggest specific synaptic aberrations that could result in a hyperexcitable, and ultimately, epileptic condition. Here we review reported changes in gene expression associated with a malformed brain, with particular emphasis on how these changes provide clues to seizure genesis.     

人睾丸与脑基因表达谱的相似性

人类的物种形成与进化问题一直是研究的一个焦点。近年来,对于人和灵长类以及果蝇等其他一些动物多种组织基因表达谱的研究表明,在人的进化过程中脑基因表达的改变最为显著,并且脑中许多基因的表达呈显著上调。信息学分析显示,在多种组织当中,人的脑与睾丸可能存在最为相似的基因表达谱。这些结果提示睾丸可能与脑类似,也在人的物种形成和进化历程中起着重要作用。本文对人睾丸和脑基因表达谱的研究进行了回顾,并提出了该研究方向今后的一些研究设想。     

Conservation of regional gene expression in mouse and human brain

Many neurodegenerative diseases have a hallmark regional and cellular pathology. Gene expression analysis of healthy tissues may provide clues to the differences that distinguish resistant and sensitive tissues and cell types. Comparative analysis of gene expression in healthy mouse and human brain provides a framework to explore the ability of mice to model diseases of the human brain. It may also aid in understanding brain evolution and the basis for higher order cognitive abilities. Here we compare gene expression profiles of human motor cortex, caudate nucleus, and cerebellum to one another and identify genes that are more highly expressed in one region relative to another. We separately perform identical analysis on corresponding brain regions from mice. Within each species, we find that the different brain regions have distinctly different expression profiles. Contrasting between the two species shows that regionally enriched genes in one species are generally regionally enriched genes in the other species. Thus, even when considering thousands of genes, the expression ratios in two regions from one species are significantly correlated with expression ratios in the other species. Finally, genes whose expression is higher in one area of the brain relative to the other areas, in other words genes with patterned expression, tend to have greater conservation of nucleotide sequence than more widely expressed genes. Together these observations suggest that region-specific genes have been conserved in the mammalian brain at both the sequence and gene expression levels. Given the general similarity between patterns of gene expression in healthy human and mouse brains, we believe it is reasonable to expect a high degree of concordance between microarray phenotypes of human neurodegenerative diseases and their mouse models. Finally, these data on very divergent species provide context for studies in more closely related species that address questions such as the origins of cognitive differences.     

High-resolution gene expression atlases for adult and developing mouse brain and spinal cord

Knowledge of the structure, genetics, circuits, and physiological properties of the mammalian brain in both normal and pathological states is ever increasing as research labs worldwide probe the various aspects of brain function. Until recently, however, comprehensive cataloging of gene expression across the central nervous system has been lacking. The Allen Institute for Brain Science, as part of its mission to propel neuroscience research, has completed several large gene-mapping projects in mouse, nonhuman primate, and human brain, producing informative online public resources and tools. Here we present the Allen Mouse Brain Atlas, covering ~20,000 genes throughout the adult mouse brain; the Allen Developing Mouse Brain Atlas, detailing expression of approximately 2,000 important developmental genes across seven embryonic and postnatal stages of brain growth; and the Allen Spinal Cord Atlas, revealing expression for ~20,000 genes in the adult and neonatal mouse spinal cords. Integrated data-mining tools, including reference atlases, informatics analyses, and 3-D viewers, are described. For these massive-scale projects, high-throughput industrial techniques were developed to standardize and reliably repeat experimental goals. To verify consistency and accuracy, a detailed analysis of the 1,000 most viewed genes for the adult mouse brain (according to website page views) was performed by comparing our data with peer-reviewed literature and other databases. We show that our data are highly consistent with independent sources and provide a comprehensive compendium of information and tools used by thousands of researchers each month. All data and tools are freely available via the Allen Brain Atlas portal (www.brain-map.org).