Exercise 16: Heat map in Prism

From BITS wiki
Jump to: navigation, search
Go to parent GraphPad Prism statistical analyses

For this exercise we will use normalized gene counts from a RNA-Seq experiment. This data set is a csv file containing a normalized gene counts for genes that were considered specific for a certain tissue showing high expression in one tissue and much lower expression in other tissues). The data set contains 6 measurements (tissues) of 6 genes.

In this exercise we will create a heat map of the data. The basic idea of a heat map is that the graph is divided into rectangles, each representing one cell of the table. The rectangle is colored according to the value of that cell.

To create a heat map you need to enter data in a Grouped table with no subcolumns. In this case, each number maps to one rectangle on the heat map. If you enter replicate values in subcolumns, you can later choose if you want the heat map to be based on the mean, median or geometric mean of the replicates. You can even choose to make a heat make of variation, and use SD, CV or SEM as the basis of the heatmap.

The arrangement of the heat map will correspond to the arrangement of the table. If your table has 3 rows and 4 columns, the heat map will also have 3 rows and 4 columns. At this moment we have genes in the columns and tissus in the rows, we would like to have that reversed in the plot: genes in the rows and tissues in the columns. To this end we need to transpose the table (= rows become columns and vice versa).

As you can see a heat map is created with default row (1, 2, 3...) and column titles (A, B, C...).

This is better but not ideal since column titles overlap, we want to rotate the column titles and place them on top of the heat map.

This is exactly what we want:


Heatmap16d.png