Automatic Annotation of Novel Genomes

As the speed at which a genome can be sequenced increases, the capacity of experimental laborotories to characterize all the proteins identified on that genome falls ever further behind. To help in bridging that gap, a certain amount of information about a protein can be inferred from its similarity to related sequences in well-characterised model organisms. In this talk, we will look at means by which the UniProtKB database transfers information from functionally characterised proteins described in the manually annotated reveived section, UniProtKB/Swiss-Prot (www.uniprot.org), onto less well studied equivalent molecules in UniProtKB/TrEMBL. Gene Ontology (www.geneontology.org), terms can also be propagated using the InterProScan tool, enabling a functional classification of the entire genome using a GOSlim. How to access tools to perform this will be described and their utility discussed.