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; } } }

time course data in microtiter plate format

choufree

can anyone help me with the graph builder to build the time course data that are recorded using a microtiter plate reader?

I have read the post about microplate map thing. While that is for 2 dimensional data. With the time data, how one can define the map role data? Thanks

Accepted answers

David_Burnham

I'm not sure you need to use the map. I would just plot my variables versus time, wrapped or filtered by vessel

Mark_Bailey

As I said before, you can use a local data filter with Graph Builder to animate the display with your time variable. Drop the data column with the well locations in the Map zone. Drop the data column with the responses in the canvas. This much is the same as the 'template.' Now click the red triangle and select Local Data Filter. Select the data column with your times and click Add. Change the interval to make it small enough to be and then either drag the interval or click the red triangle next to Local Data Filter and select Animate.

All comments

Mark_Bailey

The answer depends on what you want to do with this data. For example, if you want to examine the change in the response over time while looking at the display of the microtiter plate, then you could use animation through a data dilter, which is set to filter on the time variable, and simply add a new column for time. Drag the column with the well location to the Map zone to get the shape. Drag the column with the response to the canvas to get the mean response. Now click the red triangle next to Graph Builder and select Script > Redo > Local Data Filter. Select the column with the time for the filter.

Does this example help or are you trying to do something very different?

choufree

Thanks for your reply. My particular data analysis is for an enzyme kinetics assay.
I have a time course data in each well for 96 wells. The layout from the plate reader usually is that in the column direction it is the different time, in the row will different wells.
then the real readout data is in the table. I can draw the readout as Y and time as X in graph builder. How could you draw all the 96 sets of data as Y vs. X curve in the 96 well plate layout?

Best,

Hui

Mark_Bailey

If you want to use the microtiter plate shapes (wells) in Graph Builder, then do NOT drag the column and row to the X and Y. Instead, only drag the column with the location to the Map zone. You should then see the shapes (8 rows by 12 columns). Then drag the column with the response into the middle of the shapes.

Please see this entry in the File Exchane section of the JMP Community: Micro-Titre-Plate-Shapes

It will explain how to use them and it includes an example so you can verify that it is working as intended. You can also use the example to confirm that it will give you the result that you want.

choufree

Yes. I used this template for understanding. My application is that if you have a set of data for one well that is a time course data.
How could you draw that?

Hui

choufree

I can send you a sample data if you like.

Thanks,

Hui

David_Burnham

I'm not sure you need to use the map. I would just plot my variables versus time, wrapped or filtered by vessel

choufree

Thanks David. This could be a solution.

choufree

Hi Dave,

Do you mind showing me your data table? When you use the local data filter which column you use to do the filter?

How do you group the different column and define them as bioreators? I put the reactor name in column info. So each column is the data for this reactor. For each time point, the data is a row for a set of the reactors. In summary, I have a time series column, and all the reactors cloumns. I don't have any column to select for filter. I am stuck at this point. Thanks

David_Burnham

The data has been stacked: the first 'n' rows correspond to the time course data for the first bioreactor vessel, the following 'n' correspond to the next bioreactor vessel etc. Var is on the y axis, age or timestamp on the x-axis, bireactor is the wrapping variable and/or data filter variable.

Looking at your data table, to get it into this shape you would need to first transpose the data and then stack the columns.

choufree

Thanks a lot for the demonstration! I am almost there. Stack column is really a key.

By local data filter, I can choose certain column to show while still cann't display them individually.

As a new user, I really appreciate the help from you and Mark. Screen Shot 2017-05-12 at 10.32.47 PM.png

umberknox

Great question, visualizing time course data in a 96-well format can definitely get tricky, especially when trying to preserve both the kinetics and spatial layout of the plate.

One approach that’s worked for me is to first reshape the raw data into a long format where each row includes: Time, Well ID (e.g., A1, A2…), and the corresponding measurement. Then, in Graph Builder, you can assign Time to the X-axis and the Response to the Y-axis. To preserve the plate layout visually, use the Well ID to map the data spatially. Drag the Well ID to the Map zone, which should reflect the physical position of each well on the plate.

To see the full kinetics for all 96 wells at once, you can also consider using small multiples (facet plots). Each panel will show the time course for a single well, allowing you to detect outliers, trends, or patterns across the plate. If your software allows it, faceting by rows and columns (based on the plate layout) will make it even easier to interpret.

Also, for plate mapping and planning your visualization more effectively, you might find this 96 well plate template helpful, it’s a handy reference when aligning your data structure with the physical layout of the microtiter plate.

Hope this helps! Let me know if you need help with reshaping your dataset or setting up the small multiples.

masoncarter

A setup that usually works well is to reshape the data into a long (stacked) format rather than trying to keep one column per well or bioreactor. In that structure, each row represents a single measurement with columns such as Time, Well/Bioreactor ID, Response, and optionally Treatment or Plate ID. Then you can assign Time to the X-axis, Response to the Y-axis, and use Well/Bioreactor as the grouping, color, or wrap variable.

Ben_BarrIngh

Hi @choufree

Just wanted to add something to this conversation, another useful way to visualise plates is to combine the standard plate readout (ie Max OD per well) and the time series together with hover labels (see below)

Updating Media

To do this, you create your graphs of your time series in Graph Builder then go red button > Make into Data Table. This makes the graphs into an image for each row - bring that back over to your original table (ie with Join or Update) and then you can take the image column and use it as a label.

Attaching a data table for you to see (you need the map shape files in your maps folder).

Thanks,
Ben

Well Plate-Name.jmp

Well Plate-XY.jmp

Stacked Plate Visualisation Example.jmp