I propose an Enzyme Genomics Initiative, the goal of which is to obtain at least one protein sequence for each enzyme that has previously been characterized biochemically. There are 1,437 enzyme activities for which Enzyme Commission (EC) numbers have been assigned but no sequence can be found in public protein-sequence databases. A recent essay by | Open Access Open letter Call for an enzyme genomics initiative Peter D Karp Address Bioinformatics Research Group SRI International 333 Ravenswood Ave Menlo Park CA 94025 USA. E-mail pkarp@ Published 30 July 2004 Genome Biology 2004 5 401 The electronic version of this article is the complete one and can be found online at http 2004 5 8 401 2004 Karp licensee BioMed Central Ltd. This is an Open Access article verbatim copying and redistribution of this article are permitted in all media for any purpose provided this notice is preserved along with the article s original URL. I propose an Enzyme Genomics Initiative the goal of which is to obtain at least one protein sequence for each enzyme that has previously been characterized biochemically. There are 1 437 enzyme activities for which Enzyme Commission EC numbers have been assigned but no sequence can be found in public protein-sequence databases. A recent essay by Roberts 1 called for an effort by the scientific community to experimentally determine functions for unidentified genes in microbial genomes. Put another way the essay focused on sequences with no associated function. Here I explore the inverse problem functions with no associated sequence. I propose an Enzyme Genomics project whose goal is to find at least one amino-acid sequence for every biochemically characterized enzyme activity for which there is currently no known sequence. Roberts identifies three classes of genes whose functions would be most valuable to obtain hypothetical genes with homologs in multiple organisms conserved hypotheticals non-con-served hypothetical genes and misannotated genes. Roberts proposes that a consortium of bioinformaticians post functional predictions for these genes to a central website. Biologists would then choose candidates and test the predicted functions in the lab with results - both positive and negative - added to the same website. Roberts also proposes that the initial list of .
