Currently, if the stack height is too small to show all feature in a track, IGB draws a summary row at the top of the every track which contains Glyphs that are drawn on top of each other.

To save memory and speed up computation, not all data items are drawn and rendered into Glyphs in this top row. However, if you click-drag (rubber-banding) over the summary row to select everything that it is there, the "Selection Info" text box reports the correct number of items expected to be there.

One problem with summary row representation, however, is that it can be a bit misleading. For example, some horizontal gaps may appear, leading the user to think none of the data occupy the apparently empty location.

We need a better algorithm and representation style that will suggest to the user that:

• Individual data items are not being shown
• Any region that looks empty, really is empty (i.e., no data items from the track overlap it)

Some ideas:

• Please don't draw a histogram-style graph. The "summary" row is too short for that.
• Heatmap might be OK.
• We can show consecutive blocks with positive space and negative space only, where negative space indicates the absence of data
• We can show consecutive blocks with color intensity (or transparency) indicating the amount of data items underneath them
• Instead of drawing the summary row as a series of blocks, draw ellipses (tiny squares) in the top row wherever there is data.

In the above implementations, the summary row glyphs would not need to be selectable or countable. If a user wants to count features, they can instead set the stack height to everything being shown, collapse the track, and then click-drag over the collapsed track region.