Plant & Animal Genomes XV Conference Plant & Animal Genomes XV ConferenceJanuary 13-17, 2007
Town & Country Convention Center
San Diego, CA
Complete genome sequences have been expected to add knowledge to biology for the last decade. Annotation of a genome is an essential process to utilize the genome for further experimental and bioinformatics studies. Despite the advancement of computational methods of ab initio gene-finding, prediction of genic regions in a genome is still unresolved. Therefore, it was anticipated that full-length cDNAs (FLcDNAs) would expedite precise genome-wide annotation of higher eukaryotes. Since FLcDNAs were extensively determined in mice and humans, cDNA-based annotations played a pivotal role in these species. Recently we released the annotation of the rice genome using more than 30,000 FLcDNAs. An advantage of the cDNA-based annotation is that a gene identified by a cDNA can be examined directly by using the corresponding clone. This is in contrast to a computationally predicted gene, which is difficult to study by experiment. Another point is that an ab initio method tends to produce numerous false predictions; for example, pseudogenes are not distinguishable from functional genes. Indeed, our experience of the rice genome annotation tells that more than half of the predicted genes could not be validated by any FLcDNAs or ESTs. If convincing candidates of genes are needed, we should be cautious about ab initio gene-finding methods. However, we should also note that a substantial number of cDNAs may be missing in the current data set. Combination of ab initio and cDNA-based methods is discussed.