DESCRIPTION (Adapted from the Investigator's Abstract): High-throughput genomic sequencing efforts must be accompanied by high throughput, cost-effective sequence annotation to fully realize the value of the data. Annotation encompasses the identification and archiving of putative biological signals, sequence characteristics, and features, including genes, and, wherever cost-effective, the further characterization of those features experimentally. While one might hope that annotation could be entirely computational and thus inexpensive and rapid, computational predictions, especially of gene models, must ultimately be confirmed experimentally as an additional and independent validation of the genomic sequence data, and as a means to establish the firm foundation necessary to simplify and accelerate future biological research. The proposed work integrates computational and experimental approaches, creating a test-bed and ultimately a production system for high-throughput, high-information-gain annotation. It is designed as an open system where new computational and experimental components, and new scientific visualization tools, can be easily installed and maintained in the data management and analysis framework. Experimental annotation will be streamlined, targeted versions of standard techniques, including single pass sequencing of cDNAs selected from EST hits of genomic DNA, RT-PCR across inter- and intra-regions of putative, and dot-blots of plasmid DNA used in genomic sequencing against labeled mRNA. The three basic goals of experimental annotation are to 1) establish laboratory protocols, management structures and automation techniques for high-throughput experimental annotation; 2) validate and refine computational annotation, especially for gene model finders such as GRAIL; and 3) extract high-information-gain data, for example, by concentrating single pass cDNA sequencing efforts on ESTs from unknown gene classes, to extend the sequence similarly databases and computational gene finders. In its initial phase, development of the system infrastructure will be tightly coupled to ongoing high-throughput sequencing at the University of Oklahoma with the goal of transitioning the technology for deployment to the genomics research community at large.