| The amount of genomic data has grown massively in the past decade. Bioinformatics research has now shifted to functional genomics, which seeks to identify patterns in the data to discover new gene functions. The huge volumes of research productions from the various genomics projects make the demand for computational methods even higher. This thesis presents new methods and investigations on gene functional analysis (e.g. gene similarity, gene clustering, gene classification) using Gene Ontology (GO) and annotation databases. The proposed methods for gene similarity apply exponential function to the path length between gene annotation terms within the GO hierarchy. For evaluation, similarity values of genes taking part in various cellular pathways from the SGD database, and from other genomes were calculated and clustered. The results were compared with those from three of the leading methods. The proposed methods proved to be very competitive in all cases and the clustering results showed that our methods are able to surpass the leading methods in many cases. |
