Defining the functional landscapes of metazoan genomes

'Why are some genomes really big and others quite compact?', 'What is all that 'junk' doing in our genomes?', and 'How do limbs, fins, and faces develop and evolve?' These are 3 of the 125 Big Questions in Science, identified by Science magazine on 1 July 2005. To answer these, and other questions, requires a letter-by-letter comparison of genomes -- the books-of-life -- from very different animals. Seeing whether each letter is conserved, or else replaced, does not however tell the full story, since we also need to consider whether letters are added or deleted. Taken together, the conservation, or the insertion or deletion, of letters can highlight those parts of the books-of-life that underlie the basic biology of all animals, and others that specify the differences between species. A fourth Big Question, named by Science, is 'How much will new technologies lower the cost of [genome] sequencing?' This is beginning to be answered: sequencing will soon become so cheap that many, if not all, of the animals of interest to biologists will have their genomes sequenced by 2010. Our group has already served an apprenticeship analysing the genomes of humans, mice, rats, dogs, and chickens, with an over-arching aim of pinpointing the letters of the human genome that contribute to human health. We now wish to consider all available genomes from across the animal kingdom in order to address the Big Questions in Science. In this project, we shall find genes that encode protein, and others that do not, in each of these genomes. We then will calculate the relative contributions of protein-coding genes and non-protein-coding genes in specifying (a) the ancient biology of animals, and (b) the biology that distinguishes branches of the animal evolutionary tree. As we cannot answer all other interesting questions, we will provide comprehensive information on the internet to allow other biologists from around the world to benefit from them.

The three central issues addressed by this project are: 1. How does the total amount of functional DNA sequence vary, among various metazoan lineages, specifically arthropods, nematodes, tunicates, fish, birds and mammals, and does this variation reflect our understanding of organismal complexity? 2. Are the proportions of functional coding and non-coding sequence uniform across the metazoa, or, again does their ratio help to explain organismal complexity? 3. More pragmatically, how can we put together toolkits and data sets that will be of benefit to researchers working on species from across the metazoan clade? To answer these questions we will apply our published methods to identify, within diverse genome sequences, coding as well as non-coding sequence. These published methods are: the application of the neutral indel model (Lunter et al., PLoS Comput. Biol. 2: e5) to pinpoint sequence under purifying selection with respect to indels; prediction of protein-coding genes using transcripts from related species as templates (Heger et al., Genome Research, in press); prediction of orthology and paralogy using PhyOP (PLoS Comput. Biol. 2: e133); and, the inference of purifying selection within non-coding sequence (Ponjavic et al., Genome Res. 2007 Mar 26; [Epub ahead of print]). Although these methods could be applied immediately, we will also extend their predictive power by improving alignment quality, by clustering functional elements, and by developing Bayesian models that consider substitutions and indels jointly. Evaluations of the balance of coding versus non-coding functional sequence for very divergent genomes should improve our understanding of organismal complexity. Our studies will also shed light on lineage-specific biology of animal species that are important in agriculture and aquaculture. Our predictions of orthology and paralogy between closely-related, as well as divergent, species will be made freely available on the internet.