Figure 1

The Pancreatic Expression database web-based query interface. In this first example our goal was to identify differentially regulated genes in pancreatic cancer found in microdissected samples only and/or to extract the upstream sequence data for these genes. This query starts by choosing the "Pancreatic Expression Database" from the "Choose Database" drop-down selection in the right panel. Users will be automatically directed to the "Pancreas Expression (Build 36)" dataset with the left panel displaying the "Filters" and "Attributes" nodes with their default settings. The next step involves choosing the appropriate attributes and filters to restrict the query. Clicking on the "Filters" or "Attributes" nodes on the left will display the related page where "Filters" or "Attributes" are arranged into sections, which can be expanded/collapsed using the "+/-" box. To choose an attribute or a filter, users can simply click on the checkbox next to its description. A summary of the selected filters and attributes is automatically displayed in the left panel. Clicking on the "Count" button in the tool bar at any time when constructing the query will return the number of genes satisfying the pre-selected criteria. One can select the option to download the results to a file at the top of the result page and export them using the "GO" button. Again, there are options to change the format ("HTML", "CSV" for comma separated values, "TSV" for tab separated values, "XLS" for Excel, "ADF" for array description format) and whether to make the results unique. One can select a compressed file output and the query will run in the background to be downloaded later. One needs to provide an email address to receive an URL in a notification email that allows the query results to be downloaded. 1. Filter: in this example, the query was restricted by selecting the "Filters" node on the left and choosing the filter options on the right. Next, the filter section "GENES DIFFERENTIALLY EXPRESSED IN:" was expanded in order to select the "Pancreatic adenocarcinoma (PDAC) vs normal pancreas ND (microdissected normal ductal cells)" and limited the output to the "Pancreatic adenocarcinoma (PDAC) vs normal pancreas ND (microdissected normal ductal cells)" from the "Limit Output to" drop down menu. This is to avoid retrieving all the information about the genes satisfying the search criteria. 2. Attributes: these are arranged into five pages for "General Gene Features", "Genomic Structures", "GENE SNP Associations", "Homologs" and "Sequences". In order to select the output content, the "Attributes" node on the left needs to be clicked on and the attribute page on the right needs to be chosen. A. from the "Sequences" attribute page, the section "SEQUENCES" was expanded and "5' UTR" selected by typing in "1000" basepair as an "Upstream flank". B. from the "General Gene Features" attribute page, one can expand the attribute section "PANCREATIC EXPRESSION" where the "Gene Symbol", "Ensembl Gene ID", "Differential Expression Analysis" and "Direction of Regulation" default options are displayed. 3. Results: clicking on the "Count" button displays 1082 genes satisfying the search criteria. Clicking on the "Results" button on the toolbar retrieves the final query results. By default, this shows the first 10 results in fasta format (A) and in hyper-linked HTML format (B) but the number and format can be altered using the drop-downs above this preview, as described above. Checking the "Unique results only" in (B) displays the unique rows only from the result table.