nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Heat map, rows grouping

thomasd

Hello everyone,

Thanks for taking time to read my question.

I'll go straight to my issue with JMP:

I have several lipidomic concentration data from 10 mice in condition A and 10 mince in condition B. Whenever I try to generate the clustering analysis and the color map, JMP keeps rearranging my rows so my condition A - mice are mixed up with my condition B - mice.

I just want a heatmap with on the Y axis my 2 groups : condition A & condition B. And that each columns would correspond to a lipid.

I can use R to do this but I would really like to use JMP as it is really quick to do.

Thank you very much for your help.

Accepted answers

jules

Hi thomasd,

I believe you can get what you're looking for by placing both Subject and Dose in the "Label" section, rather than putting Dose in the "X, Grouping" section.

8196_Screen Shot 2015-03-01 at 11.53.59 AM.png

In order to get the uniform scale, check the box for "Scale Uniformly" before clicking OK.

8197_Screen Shot 2015-03-01 at 11.54.10 AM.png

Is this closer to what you want?

julian

All comments

jules

Hi thomasd,

It sounds like you're running a hierarchical cluster analysis, is that right? For that analysis output the rows are rearranged to form the dendrogram, so we'll need to do something else to form the graphic I think you want. If you're looking to preserve your row ordering and show lipidomic concentration in a heat map, perhaps Graph > Cell Plot will work for you.

Here's a simple example with the Big Class sample data set:

8185_Screen Shot 2015-03-01 at 12.40.51 AM.png 8192_Screen Shot 2015-03-01 at 12.40.58 AM.png

And here is a bigger example using the Blood Pressure sample dataset, with 4 levels of the experimental factor, and 6 DVs:

8194_Screen Shot 2015-03-01 at 12.58.23 AM.png

8193_Screen Shot 2015-03-01 at 12.56.52 AM.png

Does that get close to what you're looking for? If not, can you provide an image of the kind of plot you're looking to make? I'm sure we can find a way to make it with JMP.

Hope this helps!

julian

thomasd

Hi Julian,

Thanks a lot for your well appreciated reply.

The Blood Sample pressure dataset is actually very similar to my case. In this dataset, let's say I want to make a heatmap of the data from row 11 to row 20 (2 conditions: Control / Placebo). I want on the same heatmap the 2 conditions (so Control for the first 5 rows, and Placebo for the last 5 rows) and that each columns is a condition of Blood Pressure measurement with only one unique scale by columns for the two conditions.

Please tell me if I am not making myself clear.

Have a great sunday.

jules

Hi thomasd,

I believe you can get what you're looking for by placing both Subject and Dose in the "Label" section, rather than putting Dose in the "X, Grouping" section.

8196_Screen Shot 2015-03-01 at 11.53.59 AM.png

In order to get the uniform scale, check the box for "Scale Uniformly" before clicking OK.

8197_Screen Shot 2015-03-01 at 11.54.10 AM.png

Is this closer to what you want?

julian

thomasd

Hi Julian,

This looks exactly to what I want !

Thank you very much for your help.

I also hope it will help others.

Take care,

Thomas