Computational gene identification and function prediction by genome sequence analysis

Bioinformatics is essential for the analysis of the data generated from the human and other genome sequencing projects. The functions of large proportions of the genomes are still unknown. Although biochemical experimental procedures are necessary to pinpoint the exact function of genes, computational methods are very effective at identifying and predicting their function. A bioinformatics research group is being established that will focus on development of genome sequence analysis methods for gene identification and function prediction. Efforts will be concentrated on two major subjects: sequence similarity analysis and identification of functional non-coding RNA genes (ncRNA). We have previously developed tools for homology searches that are considerably more sensitive than those in common use today running at the same high speed. Our first aim is to further improve on these tools for detection of distant sequence relationships and also to create other tools for analysis of sequence similarity, including new concepts of presentation. Our second aim is to elucidate the role of the "intergenic" regions of genomes with focus on the computational identification and analysis of ncRNAs, whose number and importance have recently become apparent, giving raise to the field of RNomics. Verification of predictions will be done by experimental approaches.