New mutant phenotype data curation system in the Saccharomyces Genome Database

Abstract

The Saccharomyces Genome Database (SGD; http://www.yeastgenome.org/) organizes and displays molecular and genetic information about the genes and proteins of baker's yeast, Saccharomyces cerevisiae. Mutant phenotype screens have been the starting point for a large proportion of yeast molecular biological studies, and are still used today to elucidate the functions of uncharacterized genes and discover new roles for previously studied genes. To greatly facilitate searching and comparison of mutant phenotypes across genes, we have devised a new controlled-vocabulary system for capturing phenotype information. Each phenotype annotation is represented as an 'observable', which is the entity, or process that is observed, and a 'qualifier' that describes the change in that entity or process in the mutant (e.g. decreased, increased, or abnormal). Additional information about the mutant, such as strain background, allele name, conditions under which the phenotype is observed, or the identity of relevant chemicals, is captured in separate fields. For each gene, a summary of the mutant phenotype information is displayed on the Locus Summary page, and the complete information is displayed in tabular format on the Phenotype Details Page. All of the information is searchable and may also be downloaded in bulk using SGD's Batch Download Tool or Download Data Files Page. In the future, phenotypes will be integrated with other curated data to allow searching across different types of functional information, such as genetic and physical interaction data and Gene Ontology annotations.Database URL:http://www.yeastgenome.org/

Figure 1.

Database schema for mutant phenotype-related data. Top, the controlled-vocabulary tables adapted from the CHADO schema. Bottom, the phenotype data tables.

Figure 2.

Phenotype section of SGD's Locus Summary page for the ALG8 gene. The figure represents only a portion of the SGD locus summary for ALG8, http://www.yeastgenome.org/cgi-bin/locus.fpl?locus=ALG8. The observable, qualifier, and high-level experiment type are shown for all curated mutant phenotypes of ALG8. For phenotypes that include the observable ‘resistance to chemicals’, the name of the chemical is substituted for the word ‘chemical’ in this summary display; several examples of this are shown here. For the observable ‘protein/peptide modification’, the name of the Reporter is substituted for ‘protein/peptide’—in this instance, ‘carboxypeptidase Y (Prc1p)’.

Figure 3.

Phenotype details page for the ALG8 gene. Only a portion of the table is shown. Columns of the table contain: experiment type; mutant type and allele information, if any; strain background; phenotype (observable: qualifier); chemical, if any; other details, including conditions, reporters, or details; and the reference. Each observable name in the phenotype column is hyperlinked to a list of other phenotype annotations using that observable. Each chemical name in the chemical column is hyperlinked to a list of other phenotype annotations involving that chemical. This table of phenotypes is on a page that contains the standard SGD toolbar containing links to major tools and resources in SGD; below the table are links to other resources, external to SGD, which offer mutant phenotype information or provide mutant strains (data not shown).

Figure 4.

Searching phenotype data. The basic SGD Search is entered via a text box at the top of most SGD pages. Keywords entered into this search are used to search 15 major categories of information in SGD, and a summary of the results is presented on the SGD search results page (center). The link to ‘Phenotype annotations’ leads to a summary page listing the number of matches found in observables (lower left). The link to ‘Expanded phenotype search’ launches a search of all phenotype data. Results of the search are presented on the Expanded Phenotype Search Results Summary page (lower right). The top section displays matches to observables, and the bottom section displays matches to other phenotype data. On both types of results summary page, each number of matches is linked to a page displaying a table of the individual annotations.