Tabulate is not producing expected output

[#1]

Hi all

I have a bunch of data tables with genomic data in them.  Each one has a varying number of columns with the snp name as the name of the column and the nominal responses are either a 0, 1 or 2. The data type is nominal. Each row is a different patient and I have been using tabulate to create a summary of the mean of a continuous variable for each category of 0, 1 or 2.

A proper output looks like screenshot 1. 

But for some reason one data table is not putting the 0,1 and 2 as columns in the tabulate, but rather as rows under each snp.  Looks like screenshot 2.

 

 

I tried making a script of the correct one to bring into the incorrect one to manually change it but JMP crashes each time I try to do that.  The correct table script says Row Table( columns by categories... and the incorrect one says Row Table( Grouping columns...

Also, when I simply save the script from the correct one after the tabulate is complete, and then try to run it again, it simply opens an empty tabluate window, not one prepopulated with the data I used to generate the script.  And then JMP promptly crashes.

Is there a way to save a script to make a tabulate without the grouping columns format and to populate the data columns in the script?

I am using JMP 17.0.0 on a Mac computer.

[#2]

Hi @MaureenS,

Can you give us an example of what your original data table looks like including what the data type and modeling type is for the SNPs (since they are columns).

[#3 · Accepted solution]

here is a piece of it as a screenshot.

I may have found a workaround without trying to use scripts.  Funny how when you look at instructions (user guide online) you learn things.  I've always known about options under red triangles, but apparently you can right click on the column name and get more options. One is to combine tables, (columns by categories).  This did what I wanted.  Apparently the issue was lots of missing data.

Mac people don't tend to think of right clicking on things.  It's kind of crummy that there are hidden options with no indicator like the red triangle.

[#4]

Yup, that was the thing to do. Order of operations helps as well. That option would not appear if you had your traits on the top as columns then try to put your SNPs as rows. When I save script to data table and try to reopen, it gives me the same blank tabulate window and it crashes for 17.2.0. and in JMP 18.2.2 for me. But in JMP 19.0.0 coming in September, nothing crashes when saving script to data table and it is properly populated when re-running the table script.

Is it still crashing and opening as a blank table when you re-run the script/table script?