Perspectives: sequence data base searching in the era of large-scale genomic sequencing.

Large-scale sequencing of human and model organism genomes will have a profound impact on our ability to use sequence data base searching to predict the biochemical functions of sequences of interest. Despite the great value of more sequences in the data bases, a huge increase in data base size will also have adverse effects on data base searches. Upcoming problems will include (1) greatly increased search times, (2) an increase in background noise of high-scoring but biologically irrelevant matches, (3) inaccurate coding region prediction, leading to problems in protein data base searching, and (4) limited first-pass sequence annotation, making it difficult to determine the biological relevance of data base hits. Improved data base annotation tools and construction of smaller data bases of representative and highly-annotated sequences for first-pass analyses will be essential to deal with the impending flood of new genomic sequence.