Study Gene Expression and Regulation

ArrayStar’s clustering algorithms aid you in identifying genes with similar expression patterns. Parameters for both the statistical and clustering methods can be easily adjusted, allowing you to perform customized tests to capture the clearest picture of your data.

The Scatter Plot is the primary tool for a pairwise comparison of any two expression profiles. The Student’s t-test and Moderated t-test can be applied to the data in the Scatter Plot. Doing so automatically generates links showing different confidence levels and the number of genes associated with that confidence level. Links showing various fold changes and the number of genes meeting that threshold are also listed. Clicking on any of these links will select that set of genes on the Scatter Plot. The selected genes can be stored as a separate gene set within the project and further analyzed at any time.

For analysis across a series of experiments, such as a time series or a related set of conditions, two powerful clustering algorithms are available in ArrayStar: Hierarchical Clustering and k-Means Clustering.

The Hierarchical Clustering method groups data points by agglomerating them one-by-one into ever-growing groups. After grouping all of the data points, the resulting clusters are displayed in the Heat Map view with the associated Experiment Tree and Gene Tree on the horizontal and vertical axes, respectively. Clusters of interest within the Heat Map can be selected and stored as gene sets for further evaluation.

The k-Means Clustering method groups data points by partitioning them into a fixed number of arbitrary groups and then repeatedly refining the groups. By default, the resulting clusters are displayed in the Line Graph Thumbnails view. Individual clusters can then be viewed as a single full screen Line Graph for closer inspection of the results or selected and stored as gene sets for future use. In addition, one or more clusters can be selected and then re-clustered to aid in further teasing apart of potential regulatory connections.

Any clustering result can be saved and accessed later in the Clustering Result List. This view allows you to manage your clustering results, as well as choose to display your clustering result in another view.

ArrayStar also offers an Advanced Filtering tool that allows you identify genes of interest based on criteria such as: gene annotations, gene classification/ontology, signal, fold change, and statistical values. Results can subsequently be selected in all views, exported, or saved as a gene set for further analysis.

ArrayStar’s Gene Ontology view aids you in determining the biological significance of a gene selection by displaying currently selected genes in relation to a designated annotation field. Additionally, this view can be used as a tool for identifying sets of genes that are annotated with one or more ontology/classification terms.

A variety of tables are available, depending on the workflow, and columns can be added, deleted moved and sorted as desired.

The Gene Table, for example, contains detailed information for every gene in your project, including both the expression data (e.g. signal intensities and fold change values) as well as any annotations that are available from imported sources, such as the gene name(s) and Gene Ontology. Some annotations have special features, allowing you to hover over a term for more information, or click on a hot link to view more detailed information online. Any gene subsets being investigated are indicated in the Gene Table, allowing you convenient access to key tabular information for the genes being visualized by other tools in the package. The data in the Gene Table may be searched, printed, copied as a text file, or exported as a tab-delimited or comma-delimited text file.

In ArrayStar, the term “signal” is used to denote signal or regulation. Signal values can be displayed in at least one table for each of the ArrayStar workflows, as long as a signal column has been specified via the Manage Columns dialog. The following chart shows which ArrayStar tables can be used to display this information for each workflow.

To further study expression level and regulation, sets can be created from many of the views and tables within ArrayStar. You can also create Venn diagrams and dartboard diagrams to analyze gene sets.