YETI - Yeast Exploration Tool Integrator

The budding yeast Saccharomyces cerevisiae (S. cerevisiae) is one of the most widely studied eukaryotes due to its value as a model organism in biological research; it has a fully sequenced genome that is well annotated and a variety of publicly available functional genomic data sets.

Currently, there are a number of computational resources available for the visualisation and analysis of genome, gene expression and protein-protein interaction data sets individually. However, these tools tend to be focussed on a specific type of functional genomic data set and therefore do not utilise the wealth of other functional genomic data that is currently available. This is unfortunate, as many new biological insights are likely to emerge from the combined use of data from different functional genomic strategies. Despite this, there are still relatively few software tools available that are capable of bringing together a variety of functional genomic data sets and presenting them to the user for integrated visualisation, exploration and analysis. Therefore, there is a clear need for a new generation of software tools that are capable of effectively integrating the wealth of functional genomic data available for S. cerevisiae enabling users to readily utilise all of this data in their analyses and investigations of specific genes and broader biological processes.

The Yeast Exploration Tool Integrator (YETI) is a novel bioinformatics tool for the integrated visualisation and analysis of functional genomic data sets from the budding yeast S. cerevisiae. Essentially, YETI consists of two parts, a database for the storage and management of data and an associated Java program for the integrated visualisation and analysis of data. The YETI database is populated with a variety of publicly available data sets from a number of different functional genomic strategies, such as gene expression microarrays and yeast two-hybrid screens, as well as annotation data from the major yeast databases; scripts have been written to regularly update the database with the latest data from the Saccharomyces Genome Database to keep YETI an up-to-date resource.

The YETI program consists of a number of individual sections for the visualisation and analysis of specific functional genomic data sets; for example, a Genome Section for the visualisation of gene locations and a Transcriptome Section for the visualisation of gene expression data. However, these sections are closely inter-linked enabling users to swiftly move between them and investigate all aspects of any genes or proteins of interest as well as providing access to textual information, including Gene Ontology (GO) annotations, at any point. Furthermore, the YETI program utilises a group approach that enables users to collectively investigate an entire group of genes. Overall, YETI enables users to easily and rapidly explore the data in an integrated modular fashion, investigate the intricacies of broad biological processes, test specific hypotheses, and aid in the assignment of functionality to individual genes.

Analysis Section

The Analysis Section provides a sophisticated graphical interface to the YETI database with a number of different search functions to find data and an interactive data table to collectively visualise and analyse all the search results together. The interactive data table can collectively display a wide range of data on a large number of genes at once, which enables users to rapidly examine and (perhaps more importantly) compare all of the properties of an entire group of genes. The Analysis Section has a flexible and powerful QueryBuilder function that enables users to search on any aspect of the available data and perform both simple and complex database searches. One of the main reasons for constructing the QueryBuilder function was to enable users to perform keyword searches on gene names, descriptions and GO annotations to rapidly find and then collectively examine groups of related genes. Furthermore, as all the YETI sections are effectively inter-linked, once an interesting group of genes has been found they can be collectively examined in further detail in the other sections of YETI to investigate if and how they are working together to achieve their biological goals.

The Genome Section is concerned with the informative display of the S. cerevisiae genome, its chromosomes, and known and predicted genes. The Genome Section enables users to examine and compare the genomic location of multiple genes on a schematic of the entire S. cerevisiae genome. This enables users to rapidly see if a particular group of genes are all located near each other on the same chromosome or located on different chromosomes but at similar genomic positions. Furthermore, multiple groups can be highlighted and compared on the genome schematic (using distinct colours) enabling possible evolutionary relationships to be investigated. In addition, gene expression data from a microarray experiment can be overlayed onto the entire genome schematic enabling users to visualise the expression profile of the entire genome and rapidly find regions of co-ordinated expression.

Chromosome Window

Any of the 16 nuclear chromosomes can be individually examined in much more detail in the Chromosome Window. This window enables users to easily scroll along the whole chromosome to view the location and distribution of genes and rapidly find areas of interest. Once an interesting region has been found, such as the region surrounding a particular gene or feature of interest, it can simply be selected by dragging the mouse to create a selection box. All the selected genes can then be collectively investigated in further detail in the other sections of YETI; for example, the Analysis Section would display a wide range of data on all the selected genes enabling users to examine if they are involved in the same or related biological processes. Alternatively, a single datasheet on any of the individual genes can be viewed simply by mouse clicking on them. An advanced feature of the Chromosome Window is the ability to overlay one of the individual microarray experiments currently stored in the YETI database onto the chromosomal display. This is accomplished by colouring each gene box with a colour that reflects its expression and therefore enables users to rapidly find regions of co-ordinated expression along the chromosome.

Transcriptome Section

The Transcriptome Section provides an effective means for the visualisation and analysis of gene expression data. At the heart of the Transcriptome Section is the graphical panel which can display any one of the hierarchically clustered gene expression microarray data sets currently stored in the YETI database; these data sets are normalised and clutered by the Cluster program before database importation. The genes are ordered with respect to the data set's hierarchical tree which is also displayed on the graphical panel so that the relationship between genes can be easily examined. The graphical panel is contained within a scrollpane which enables users to easily scroll up and down to examine the entire hierarchically clustered gene expression data set and rapidly find regions of interest such as a particular expression cluster. Any region of interest can simply be selected by dragging the mouse to create a selection box enabling users to collectively investigate all of the genes located in this region in the other sections of YETI; for example, the Genome Section would highlight the location of all the selected genes on the genome schematic enabling users to examine if they are located in the same or similar regions. The YETI database is currently populated with two large microarray data sets, specifically Gasch et al. (2000) and Gasch et al. (2001).These two studies were chosen as they are large genome wide data sets that complement each other well, monitoring how S. cerevisiae cells respond to a wide variety of environmental conditions and DNA damaging agents. The Gasch et al. (2000) study is especially well respected and as a result is commonly used as a test data set for gene expression analysis programs;

Proteome Section

The Proteome Section is concerned with the effective visualisation of protein-protein interactions on a dynamic graphical display panel. The graphical panel uses a 'springs and rings' type relaxation algorithm to automatically arrange all the displayed proteins in an optimal way; this algorithm is based on the publicly available relaxation algorithm from the Sun network mapping Java applet but was inspired by Mrowka, 2001. All the protein-protein interactions of a specific protein of interest can be visualised simply by entering the protein's name; YETI then searches the database for all interactions involving the selected protein and displays any interactions found dynamically on the graphical panel. A datasheet on any of the displayed proteins can be viewed simply by mouse clicking on them. Alternatively, multiple proteins (such as all of the proteins in a particular interaction cluster) can easily be selected and then collectively investigated in further detail in the other sections of YETI; for example, the Transcriptome Section would highlight the location of all the selected proteins corresponding genes in the hierarchical tree enabling users to examine if these interacting genes/proteins are located in the same expression cluster. The Proteome Section is also able to display all of the interactions of an entire group of proteins; for example, the Analysis Section could be used to search for all the proteins involved in a specific biological process and the Proteome Section would highlight and display all of their interactions enabling users to investigate how they are working together to achieve their goals and what other proteins they may be working with. An advanced feature of the Proteome Section is the ability to overlay one of the individual microarray experiments currently stored in the YETI database onto the display panel. The YETI database is currently populaed with over 20,000 interactions from the GRID database.

The Function, Process and Component (FPC) Section is concerned with enabling users to browse GO annotations and define specific groups of genes which can then be investigated in further detail in the other YETI sections. At the heart of the FPC Section is the GO annotation list which contains all the GO annotations that have been used to characterise the genes of S. cerevisiae. Single or multiple annotations of interest can simply be selected by mouse clicking on them and then assigning them to either the Red or Green group. Therefore, the FPC Section enables users to easily and rapidly construct specific or broad groups of functionally (or spatially) related genes to collectively investigate in further detail in the other sections of YETI. For example, the FPC Section could be used to select all of the genes characterised with the 'mRNA splicing' annotation and the other sections of YETI could be used to investigate the dynamics of how these genes are working together in order to achieve their biological goal and also investigate what other genes they may be working with, possibly leading the characterisation of previously uncharacterised genes. Furthermore, as two different groups of genes can be defined this enables the properties of both groups to be collectively examined and compared.

Datasheet Window

The Datasheet Window displays a wide range of information on a single gene of interest (such as its names, length, number of introns, chromosomal location, phenotype, descriptions, and GO annotations) and can be launched from numerous points within the YETI program. Overall, the Datasheet Window enables users to easily and rapidly view a wide range of information on a specific gene of interest. However, the Datasheet Window also contains a number of advanced options that provide direct links to the other sections of YETI where data relating to the selected gene is automatically displayed and highlighted. For example, the Transcriptome Section would highlight the gene's location in the hierarchical tree or the Analysis Section would display a wide range of data on all the genes that share the same GO annotation. These links are especially useful when investigating a potential function for a gene of unknown function as it enables users to investigate what other genes the selected gene might be working in through its chromosomal location, expression and interaction properties in a 'guilt by association' approach. Furthermore, these links are also useful when investigating what other genes a gene of known function may be working with in order to achieve its biological goal.