Investigating Unsequenced Enzymatic Activities: A Preliminary to Enzyme Genomics

Sri International, Menlo Park CA

Investigators

Abstract

An analysis of the contents of the ENZYME database has revealed that prima facia, no gene or protein sequence is known for more than 1400 enzyme activities, corresponding to 38% of enzyme classification (EC) numbers. This lack of sequence data for such a large fraction of enzyme activities hinders research and biotechnology in a number of areas ranging from genome annotation to metabolic engineering and pathway prediction. Fortunately, the recent availability of large numbers of completely sequenced genomes enables the eventual identification of the genes encoding these enzymes using data available in the literature, combined with computational and experimental analyses. The overall goals of the project are to perform a survey of the literature associated with orphan activities to further substantiate that their existence is not artifactual, and to capture data from this literature, which would facilitate the identification of the genes encoding these activities. Thus, the project will 1) authenticate orphan activities; 2) capture and disseminate data that could enable the identification of the genes coding for enzymes associated with orphan activities; and 3) submit any sequence data found in the literature to the UniProt database. The survey will be applied to a randomly selected subset of orphan activities large enough to provide reasonably solid conclusions relative to the universe of orphan activities. Intellectual Merits Assessment of the authenticity of orphan activities: Determining an upper bound on the rate of artifactual causes for orphan activities will inform the decision as to whether to proceed with an Enzyme Genomics Initiative. Similarly, extracting molecular properties and other data from the literature associated with orphan activities will help determine the extent to which such an Initiative is warranted. Enhanced genome annotation: The availability of gene and protein sequences encoding previously orphan activities will enhance our ability to annotate genomes in terms of both coverage (fraction of genes that can be recognized), and accuracy (fraction of predicted gene functions that are correct). Enhanced pathway prediction: The availability of sequence data from previously orphaned activities will also increase our ability to predict computationally the metabolic pathway component of organisms, since such predictions typically rely upon sequence data from known enzymatic activities. Enhanced metabolic engineering: These sequence data will also enhance the practice of metabolic engineering, again because of its dependency upon sequences from known enzymatic activities. Broader Impacts Increased database accuracy: The outcome of this work will result in the expansion of enzyme sequences in the UniProt Database, a major protein information database. Increased value of existing data: The results of this work will add value to a large body of enzymology whose full value is not realized currently because of the absence of sequence data, and will further leverage systematic genome sequencing.