Multi-Dimensional Projections

You can project each individual cohort onto two or three principal components, creating a two-dimensional or three-dimensional projection. You may generate more than one analysis at a time.  For more information, see Asynchronous Operations.

Note:  To display a three-dimensional projection, you must have Java 3D installed.

To view a multi-dimensional projection of principal components:

1.  Run tranSMART, then click the Dataset Explorer tab.

2.  Select the study of interest.

3.  Define the cohorts whose data points will be represented in the PCA visualization.

4.  Click Advanced > Principal Component Analysis.

5.  In the Compare Subsets-Pathway Selection dialog, specify the platform other variables, or accept the defaults.

6.  Click Run Workflow.

The PCAViewer appears.

7.  Click the Projection tab:

8.  Select the two or three principal components to use in the visualization.

9.  Click the Plot button.

The following figure shows a two-dimensional projection for the first and second principal components.  The two different colors in the scatter plot represent the different subsets.

Note that hovering the mouse pointer over a marker in the scatter plot displays the data that the marker represents.