Multiplex MSD 96 Well Plate with Spots Map File

[#1]

I have used your files and created map file for MSD 96 well multiplex plates with 10 spots per well. I have named the 10 spots "1" to "10" and the well as a whole "11" all in the Part ID column. The Shape ID's are all still the 96 wells using the LetterNumber notation.  This looks GREAT, but when I go to use it, speficying Well as for the Map Role, I only color the entire well, not the individual spots within each well.  How do I access the Part ID portion of the shape file in the Map Role?

Thanks,

Martin

@Mark_Bailey 

[#2]

Graph Builder uses shape files to render data with a geometric interpretation. The shapes are defined by two JMP data tables (Name and XY). JMP published the specification for the required format of these files. JMP installs with many pairs of shape files for geographical entities. But you can make your own shape files for any custom geometry. You just need to follow the specification and it should work without any problem.

I used the specification to make the shape files for standard 96-well plates. The specification provides the means for each shape to be comprised of more than one part. So you could make your own shape files with 10 parts (spot) per shape (plate well) to display your nested data.

[#3]

So the shapes that are referenced in your data table are based on the second column in the Name file, right?  This is a clip from my Name file:

but what I have done is add 10 additional Part ID's (one for each spot within each of the 96 wells) for each Shape ID in my XY file.  So for each Shape ID in the XY file, I have 11 Part ID's (10 for spots and 1 for the outer circle).  My final graph looks like this:

In order to use the individual spots, instead of the outer 96 wells, how do we use the combination of Well Number and Spot Number?  I thought it was with different Part ID's, but can't seem to use them.

[#4]

I apologize. I made a mistake. The map of a response value to a shape is key. So you should not use the part feature for each spot. Each spot should be its own shape. Then the mapping will work.

[#5]

Ok. What is the Part ID used for then? Only for non-contiguous areas, e.g. the US with the continental US, Hawaii and Alaska all being part of the same country?

So Shape ID would represent country, and Part ID would just be the non-contiguous areas?

Thanks for the help.

[#6]

Yes, that is the idea. But you do not have dis-contiguous areas. Part ID will always be 1.

The shape corresponding to a single response is a spot in the well. Each shape has presumably just one part. This one-to-one mapping is the same one that I used for the original plate well.

[#7]

If I understand you @Mark_Bailey, Each of my 10 spots within each of the 96 wells will get its own Shape ID (960 in all). This means I need to give each shape a unique name in the Name file. Normally, my analyses have two columns: one for Well, e.g. B5 or H10, and one for spot, e.g. 1 or 8. I guess I will need to make a new column in my analysis table that uses a Well/Spot identifier and I can match up with the map Name file second column. Is that correct? Or, is there a way to have two columns in the map Name file represeting Well and Spot that can be used in the map role in my analysis table?

[#8 · Accepted solution]

Yes, that is correct. There is one column that uses the same name as the one used to define the Name and the XY file. You might make an easy-to-use scheme based on concatenation. Easy for you (Well and Spot), easy for JMP (Location). For example:

So the location value is the name used in the Name and the XY shape files. But the scheme is entirely up to you. I tried to make my original scheme as straight-forward and easy as possible because I intended to release it in the JMP Community.