High-content screening is essentially automatic microscopy. After images are acquired, processes of image segmentation and feature extraction are applied to them to generate lots of numbers. A numerical data set resulted from HCS is generally big and messy because of the nature of HCS and the heterogeneity of cell populations.

The following features are especially designed for HCS:

• Cell plot
• Plate heat atlas
• Plate image atlas
• In situ zooming
• Cell cycle analysis
• Cell gallery
• Initiating data and image loading directly from Cellomics Store.

All these HCS features are full integrated with Panmo's other features to form a synergistic environment, under which HCS information just flows at your fingertips while you are exploring your HCS data, as illustrated by this screencast.

Cell Plot

A cell plot is sort of like a scatterplot with cell images replacing circles or dots. Examples are:







Like scatterplots, cell plots support logical zooming, too.


Top

A plate heat atlas displays a collections of tabbed pages of heatmaps and tabbed pages of image trellises, as illustrated by the following animation (if you don't see animation, you'll have to adjust your web browser to show animation.)



For each heatmap in a heat atlas, the minimum well summary is printed at the lower left corner and is always left-adjusted to the left edge of the plate heat atlas. The maximum well summary is printed at the lower right corner and is always right-adjusted to the right edge of the plate heat atlas.

To get a closer look at the scan image in a panel of an image trellis, click the left mouse button over it to pop up a window displaying an enlarged version of the scan image. This window also serves as an in situ zooming controller.

When you move the cursor over a heatmap, Panmo will continuously display information on the well underneath the cursor at the bottom of the plot. For example, the cursor is over the B3 well in the following plot:



With a few mouse clicks, different colormaps can be used.





A bit of explanation about the colormap used in the last plate heat atlas. This colormap only has 9 colors. The numbers to be encoded by these 9 colors are sorted into ascending order. The smallest one third of the numbers are assigned shades of blue. The middle one third of the numbers are assigned shades of green. The largest one third of the numbers are assigned shades of yellow. So the median value is always assigned the color green, which is the color pointed at by the black-and-while pointer in the last plate heat atlas.
Top

A plate heat atlas is closely related to plate heat atlas. When all heatmap pages are zapped in a plate heat atlas, we are left with a plate image atlas. Panmo has a command to make plate image atlas directly and quickly.


Top

All images in plate image atlases and plate heat atlases support in situ zooming:
1. Magnification power less than 1: The window displaying a single image below is an in situ zooming controller, which is derived by clicking the left mouse button over the C1 well in the image trellis. The yellow rectangle is the area we zoom into. All images (be it B&W or composite) in a plate heat/image atlas share the same yellow rectangle. There can be more than one in situ zooming controller around at the same time for a plate heat/image atlas. The yellow rectangle can be moved around or resized easily.
   
   
   
   
2. Magnification power equal to 1:
   
   
   
   
3. Magnification power greater than 1:
   
   

Top

Examples of cell cycle analysis menus:





The plate map in the "Control Wells" page is for selecting control wells. Control wells are used to compute cell cycle phase boundaries. Once we have phase boundaries, data in all wells are classified into one of the cell cycle phases and the result can be displayed as a barplot, a trellis of barplots, a pie chart, a trellis of pie charts, a DNA profile, or a trellis of DNA profiles.
Top

All images of cells in a cell gallery are displayed at the same scale so that they are comparable to each other. Users can choose the channels to be displayed and if composite images are displayed or not. The ordering of images can also be determined by a variable (such as ObjectTotalIntenCh1).

A cell gallery example:



To get a closer look at any one of the images, just click the left mouse button over it:


Top

The dialog lists all plates initially. As you enter more and more criteria, the number of plates listed in the scroll area will decrease. No need to hit the Return or Enter key; the list of plates in the scroll area will be updated continuously as you type.

You can tell the system to load images only.
Top