Basic Expression and Exon Workflows in JMP Genomics 3.2
This blog post is the first of two that will describe new basic workflows in JMP Genomics 3.2. The basic workflows offer an intermediate step between the highly flexible Workflow Builder first introduced in JMP Genomics 3.0 and the simplicity of the streamlined Affymetrix Expression CHP Wizard, which first appeared in JMP Genomics 3.1.
Both the CHP wizard and the Basic Workflows take advantage of the Workflow Builder infrastructure, but differ from using the Workflow Builder directly. When working within the Workflow Builder dialog, you first need to open other JMP Genomics process dialogs, select parameters, then save those settings to a workflow, or select from pre-existing saved settings for various processes to build up a workflow. Because the Workflow Builder can incorporate settings for any JMP Genomics process that runs SAS, it provides a huge range of options for customizing data processing and analysis steps. However, it can be intimidating for a new user to get started using it.
The interactive CHP wizard and the dialog-based Basic Workflows both build settings for you behind the scenes and launch workflows automatically. The CHP Wizard and the Basic Workflows provide a restricted set of options, but you can modify or build onto the workflow after its initial run. The Basic Expression and Basic Exon Workflows offer the option to generate several standard quality control plots, with the choice to display these for data before and after normalization, and a tab to specify ANOVA model and options. The Basic Exon workflow automatically creates the appropriate model to test for alternative splicing.
New Action Buttons launched with the ANOVA results to enable you to quickly drill down on interesting findings. In the case of exon data, you can drill down on a transcript cluster ID to see box plots and line graphs to reveals which probesets (exons) appear to display tissue-specific differences. You can also test the statistical significance of the intensity differences for individual exons between groups (e.g., control vs. treatment).
The output for all the workflows is a JMP journal with script buttons to launch output from each process – a very convenient presentation tool. In fact, our lead developer Russ Wolfinger recently demonstrated results from the Basic Genetics Workflow and Basic Copy Number Workflow at the MAQC II face-to-face meeting at the FDA in Rockville, MD. These results were run on CHP and CEL files, respectively, derived from a subset of the Wellcome Trust Case Control Consortium data set. Like the JMP Genomics predictive modeling processes, which have gotten quite a workout from the team’s involvement in MAQC I and MAQC II, these new workflows will certainly be stretched and shaped by the JMP Genomics team’s involvement in MAQC.
We have found these basic workflows to be extremely useful in our own research, especially when beginning with a new data set. Please give them a try and send us feedback once you have JMP Genomics 3.2.