We have used the annotations of six animal genomes (Homo sapiens, Mus musculus, Ciona intestinalis, Drosophila melanogaster, Anopheles gambiae, and Caenorhabditis elegans) together with the sequences of five unannotated Drosophila genomes to survey changes in protein sequence and gene structure over a variety of timescales--from the less than 5 million years since the divergence of D. simulans and D. melanogaster to the more than 500 million years that have elapsed since the Cambrian explosion. To do so, we have developed a new open-source software library called CGL (for "Comparative Genomics Library"). Our results demonstrate that change in intron-exon structure is gradual, clock-like, and largely independent of coding-sequence evolution. This means that genome annotations can be used in new ways to inform, corroborate, and test conclusions drawn from comparative genomics analyses that are based upon protein and nucleotide sequence similarities.

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' (E. G. Miller et al., 2000), 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.

The nucleotide sequence of the Drosophila melanogaster genome is now available and efforts are ongoing to improve its accuracy and annotation. The value of these sequence data will be enormously enhanced if we can provide for each gene information on: (1) the structure and expression pattern of its transcripts; (2) the mechanisms and logic used to control their expression; and (3) the functions of their protein products. My presentation will be directed toward the challenges of obtaining and interpreting such information for the fruit fly. Gene sequence and expression pattern databases will be extremely powerful tools. However, the function of a protein in a multicellular organism depends on context and will almost certainly need to be determined by experimental analysis. Collection of the required large datasets will be difficult and neither the intellectual framework nor experimental tools for analyzing complex gene networks are currently in place. Nevertheless, there is reason for cautious optimism that the complete genomic sequence of organisms will enable the necessary global approaches to study gene function and regulation. The conservation of gene structure and function during evolution will allow for the linking and sharing of information garnered in different experimental systems. But what data should be collected and how to interpret these data are much less clear. I will describe recent attempts by the Berkeley Drosophila Genome project to address some of these issues.