Workflows provided by SmartR:
- Boxplot Workflow
- Correlation Workflow
- Heatmap Workflow
- IPA Connector Workflow
- Linegraph Workflow
- PatientMapper Workflow
- Volcanoplot Workflow
In addition to the Advanced Workflows a more interactive mode of visualising data in tranSMART has been created. SmartR uses modern technologies to create interactive graphs directly from within tranSMART. Although the technologies are different, some of the functionalities are similar. This means a user will have multiple options for creating a desired graph. Which of the two technologies is best depends on the type (and sometimes the volume) of data to visualise.
How to run SmartR workflows
To begin to run any analysis:
In Analyze, open the study of interest.
Define the cohort(s) you want to analyze by dragging one or more concepts into empty subset definition boxes. For more information, see Defining the Cohorts.
Then, start a SmartR workflow:
Select a SmartR workflow from the main SmartR tab.
Empty selection boxes appear that allow you to drag in numerical, categorical and high dimensional nodes of interest. The boxes change colour based on whether suitable input is detected:
For most workflows you need to select specific markers from the high dimensional data set:
Click Fetch data. This will transport the data from tranSMART into the SmartR computational R environment. Once ready, SmartR will provide summaries of the retrieved data.
(Optional) The pre-processing tab allows you to perform modifications to your data if this is necessary. For instance, this could be recalculating z-scores based on current selection criteria, or performing probe aggregation.
Use your data in SmartR analyses by clicking Run Analysis. The page you see there is unique to the workflow you have chosen. The following sections describe how to run specific analyses after you perform the above steps.
If you want to rerun a workflow after changing the source data, you always have to click Fetch data again.