HCS Tutorial 4

Data:

This data set is from an siRNA transfection experiment, in which the TNF-R (Tumor Necrosis Factor Receptor) is knocked down by siRNA.

To follow the steps in this tutorial, you are welcome to request a copy of the trial edition of Panmo and a data set very similar to the one used here.

Assay:

Cells are first transiently transfected with siRNA specific to TNF-R.

There are a bunch of downstream effectors in the TNF signaling pathways. We are only looking at 2 of the downstream effectors; hence, a much simplified diagram is drawn here. One target is NF-kB, which moves from cytoplasm into nuclei when it becomes activated. The other target in this experiment is ATF2, also a transcription factor. It also moves from cytoplasm into nuclei when stimulated.

Cells are then treated with TNF for 0, 15, 30, and 45 minutes before being fixed and processed for examination.

Readout:

Three cellular parameters are monitored:

Channel 1: DNA stain (for identifying nuclei).
Channel 2: NF-kB. We are interested in its cytoplasm-to-nucleus translocation.
Channel 3: ATF2. We are interested in its cytoplasm-to-nucleus translocation.

Numbers of Cells in Wells

Let's first take a look at the numbers of cells in the wells and paint those cells untransfected with TNF-R siRNA green and those cells transfected with TNF-R siRNA red. Here are the steps:

[ Note: The following steps are captured in this 24-second screencast. ]

Invoke Browse in the primary console:

and select Well in the ensuing menu:
After clicking the O.K. button in the above menu, we get a barplot:

The Y axis in the barplot is the number of cells.
We'll make all cells in all well green first. Click the Color button in the primary console:

and select the green color in the ensuing color palette:

The barplot of Wells becomes this

after we click the O.K. button in the color palette.
Next we are going to paint cells in wells A2, B2, C2, and D2 red. The above barplot is already in paint mode and the color of the paint brush is red. When we move the cursor over this barplot, the cursor will change to a paint brush. To actually paint anything, we have to press and hold down the left mouse button to get a rectangle. The color of the rectangle is the current paint color. We just drag this rectangle around; anything touched by this rectangle will be painted. This is what the barplot looks like after we paint cells transfected with TNF-R siRNA red:

The 2nd Channel, NF-kB

We'll take a look at the cytoplasm-to-nucleus translocation of NF-kB in cells in all wells. Here are the steps:

[ Note: The following steps are captured in this 13-second screencast. ]

Invoke Browse in the primary console

and make the following selections in the ensuing Browse menu
Clicking the O.K. button in the above menu will get us a boxplot.

It is clear that NF-kB becomes activated quite early on after TNF treatment. Its level in nuclei stays high even after the 45 minutes of treatment. We can easily see here that the nuclear translocation of NF-kB in those cells transfected with the siRNA is reduced quite a lot.

The 3rd Channel, ATF2

We'll take a look at the cytoplasm-to-nucleus translocation of ATF2 in cells in all wells. Here are the steps:

[ Note: The following steps are captured in this 14-second screencast. ]

Invoke Browse in the primary console

and make the following selections in the ensuing Browse menu
Clicking the O.K. button in the above menu will get us a boxplot.

The level of ATF2 in nuclei also jumps up quite fast but obviously its increase in nuclei is more transient, compared with NF-kB. We can see that its level in nuclei goes down quite a lot at the 45 min time point.

Point to Ponder

Since siRNA transfection is not homogeneous, does it make sense not to examine subpopulations of cells with altered responses?

We all know that the transfection efficiency in siRNA experiments is never 100%. That is why people call it siRNA knockdown, not siRNA knockout. Let's take a look at the boxplot of NF-kB translocation again. We can see that the NF-kB translocation of red cells in wells B2, C2, and D2 is suppressed. but it is never as low as that of untreated cells in A1 and A2. Why is that? Because now wells B2, C2, and D2 contain mixed populations of cells. Some cells still respond normally to TNF but some don't. Those cells that don't respond to TNF normally have their TNF receptors knocked down more. Can we examine those cells to see what happened to their other downstream effectors? Sure, with Panmo, it is quite easy to do so.

Exploring subpopulations of siRNA-Transfected Cells

Here we are going to visualize the data in the boxplot of NF-kB translocation from a different viewpoint by making a trellis graph of histograms. After that, we'll take out all the siRNA-transfected cells in a well and explore how ATF2 was affected by the siRNA and how NF-kB and ATF2 relate to each other.

First, to make a trellis graph of histograms of NF-kB translocation, take the following steps:

[ Note: The following steps are captured in this 20-second screencast. ]
1. Invoke Histogram in the primary console:
2. Make the following selections in the ensuing histogram menu:
3. Click the O.K. button in the above menu to get a dialog for setting trellis display parameters:
  
  Panmo remembers the settings of Panel Order, Strips?, Packet Size Bars?, and Abbreviation Threshold used last time. If you are actually running Panmo to follow this tutorial step by step, what you see on your computer might not be the same as this screen shot. Adjust the settings of these 4 parameters to be the same as those shown here; otherwise, you won't get the same trellis graph as this tutorial will get later.
4. We move Well from the attributes column to the selected attributes column by
  - Drag and drop, or
  - Click the left mouse button on it and press the S key on the keyboard.
  After we do that, this trellis menu will become:
  
  Note that Well is printed on a light yellow background because it is a categorical variable. Numerical variables are printed on a light blue background. Also note that the number of columns and the number of panels are updated to 8 automatically because Well has 8 different categories.
5. We next adjust the Columns and Rows fields to make all the 8 panels line up in a column of 8 rows. This way, it's easier to compare the distributions of NF-kB translocation in different wells.
6. After clicking the O.K. button in the above dialog, we get the following trellis graph:
Next we are going to take out all the cells in the B2 well. All cells in this well were transfected with this TNF-R siRNA. 2 steps to achieve this:

[ Note: The following steps are captured in this 10-second screencast. ]
1. Put the above trellis graph in the trellis display mode:
2. Move the cursor over to the Well: B2 panel and click the left mouse button to get this histogram panel plot:
Now we have all the cells in the B2 well. Let's divide them into 2 groups: those cells with low NF-kB translocation and those cells with high NF-kB translocation. We'll do so with Panmo's paint brush. NF-kB low responders will be painted blue. NF-kB high responders will remain red after the painting operation.

[ Note: The following steps are captured in this 13-second screencast. ]
1. There are more than one way to change the color of Panmo's paint brush. We'll invoke a keyboard shortcut to save time. Press the B key on your keyboard to get a blue paint brush:
2. Proceed to paint the bars in the above histogram panel plot:
  
  For those of you not actually running Panmo to follow this tutorial step by step, here is a screen shot of the computer screen up to this point.
Let's compare the ATF2 translocation of NF-kB low and high responders in the B2 well. Because the above histogram panel plot contain all the cells in the B2 well, we just get all the cells from it and make a boxplot with Observation color (blue for NF-kB low responders; red for NF-kB high responders) as the X variable and ATF2 Translocation as the Y variable:

[ Note: The following steps are captured in this 19-second screencast. ]
1. Put the above histogram panel plot in the data retrieve mode by clicking the left mouse button over
2. Click the right mouse button over the histogram panel plot to pop up its right-click menu:
3. After we invoke Retrieve Displayed Data, the primary console will disappear and a temporary console will pop up to contain all the cells in the histogram panel plot. We invoke Browse in this temporary console:
4. Select Observation color and ATF2 Translocation in the ensuing browse menu:
5. After clicking the O.K. button in the above browse menu, we'll get a boxplot:
  
  It's clear from this boxplot that those cells with low NF-kB translocation also show low ATF2 translocation.
Next we are going to explore how the ATF2 translocation relates to the NF-kB translocation in B2 cells. The histogram panel plot of NF-kB translocation in B2 cells is already in the data retrieve mode. We just use it as our starting point, retrieve all the displayed data to put in a temporary console (Retrieve Displayed Data), take a look (Browse) at NF-kB translocation and ATF2 Translocation, and get the following scatterplot:

Note that we have resized this scatterplot to make the horizontal scale roughly the same as the vertical scale. We can easily add a lowess curve to this scatterplot to provide a visual abstraction of the pattern in the point cloud. Do so by invoking Add Lowess Curve in the right-click menu of the scatterplot. The following is the above scatterplot with a lowess curve added:
We can easily specify what to explore with Panmo. In the last example, we'll demonstrate pooling the data in the B1 and B2 panels of the trellis graph of histograms and examining the relationship between NF-kB and ATF2.

[ Note: The following steps are captured in this 34-second screencast. ]
1. First we put the trellis graph of histograms in the data retrieve mode:
2. Next we invoke Retrieve Data by Rubber Band... in its right-click menu:
3. At this point, the cursor will switch to . We just move this cursor to the place that will be one corner of the bounding box, press the left mouse button, and drag the bounding box into shape. The following snapshot was taken right before the left mouse button was released at the place to the left of the 3 light blue dots. The left mouse button was first pressed down at the place to the right of the 3 light green dots and was dragged toward the 3 light blue dots. These dots were added to this snapshot to identify the starting and the ending place of the drag.
4. Once after the left mouse button is released, all the data within the bounding box (cells in wells B1 and B2) are collected and put into a temporary console. From there, we just proceed to Browse NF-kB Translocation and ATF2 Translocation to get the following scatterplot: