Contents - Index

Heatmap Analysis

(This feature is only available in GenEx Pro/Enterprise)

Heat maps are a way of visualizing hierarchical agglomerate clustering, which is the most common method for grouping data. The construction of a hierarchical agglomerative classification can be achieved by the following general algorithm.

1. Find the two closest objects and merge them into a cluster.

2. Find and merge the next two closest points, where a point is either an individual object or a cluster of objects.

3. If more than one cluster remains, return to step 2 

The clustering algorithm is used to group rows (samples) and columns (genes) in the active data file to produce two cluster trees. The samples and genes are arranged in the order given in the tree so that each data row in the data file corresponds to a column in the heat map and vice versa. The color of a cell in the heatmap relates to the expression of the gene in the sample that the cell corresponds to. 

Open the Cluster tab among the analysis tabs in the top of the main window, and press the Heatmap analysis button to load the analysis into the Control panel.

By pressing the Run button, the clustering is performed with the default settings which is average linkage as clustering method and  Euclidean distances as the distance measure. Press the Advanced button to select a different clustering method or distance measure. Both dimensions in the data can independently be analysed according to separate clustering methods and distance measures. 

The following clustering methods are available.

• Average linkage
• The complete linkage,
• The single linkage
• Ward's method 

The distances between objects can also be measured differently. Most common for continuous data, where we measure gene expression in copy number or Cq values, are the following.

• Euclidean distance
• Euclidean squared distance
• Manhattan distance

The following are used to measure the distances for discrete data.

• Dice coefficient
• Jaccard coefficient 

You can also use the following as distance measurements for both continues and discrete data.

• Pearson correlation
• Pearson magnitude.

A resulting heatmap will look something like this in GenEx:

Note that nodes in the cluster trees (yellow dots) are clickable and that clicking these nodes will switch the orientation of the branches in a way that will allow you to customize the heatmap according to your own specific desire.