Gene prediction and functional annotation in the Vitis vinifera genome

In the last years the increasing number of sequencing projects and the availability of completely sequenced genomes pose the problem of searching for gene sequences in a rapid and reliable way. Bioinformatics is playing a fundamental role in this research field. In fact, many bioinformatic tools and software that consider multiple and heterogeneous evidence sources have been developed in order to improve the genome annotation. Genome annotation can be divided in two distinct phases: gene prediction and functional annotation. The prediction phase is the process to identify the exact gene structure, delimiting the exon-intron boundaries and the localization of genes on the genome. Otherwise, the functional annotation is the action of characterizing predicted genes, assigning them a biological function, a metabolic role or describing structural features. This PhD project focuses on the development of computational methods for the management of data coming from a genome sequencing project. The work consists on the implementation of a bioinformatic platform for gene prediction and functional annotation of the Vitis vinifera genome. This work has been carried out in collaboration with CRIBI bioinformatic group, that is member of the Grape sequencing project. The annotation platform consists of two distinct modules. The first module regards gene prediction. Different computational methods showed a great reliability to discover molecular signals and to reconstruct gene boundaries, becoming fundamental in the annotation at genome-level. These methods are represented by ab-initio predictors, genome alignments of ESTs or proteins or comparative genomics. Otherwise, in the second module of annotation platform, the predicted genes are functionally characterized, adopting mainly a similarity approach. This approach bases on the assumption that regions highly conserved maintain the original functions or roles also in different species. This project includes also the development of databases and tools to store and retrieve genome data. In particular, the PhD work focused on the implementation of a XML-based query system that permits the information retrieval through web page access and, in the next future, also through web-services workflows.