Parameters | Genetics | Correlation Value for Clustering

Use this feature to specify a correlation value between 0.7 and 1 to use to determine the minimum radius when Automated Radius K Means is selected as the K-Means Clustering Method.The higher the correlation, the more clusters that are defined, and the more highly members of each cluster correlate with one another. Denoting the correlation that you specify as Rho, the radius for SAS PROC FASTCLUS is set equal to sqrt(2 (Mean_USS)(1 - rho)), where Mean_USS is the average of the columnwise uncorrected sums of squares.Note: This option is available only when the Use K-Means clustering to reduce number of markers check box is checked.

Tip: To change the scale of the slider, right-click on the slider and select from the pop-up menu. Change the upper and/or lower boundaries in the window that appears, to rescale the slider and click .