Overview: I believe it is an issue with some of the initial triangles since newer ones don't always exhibit this behavior? Maybe a integer comparator used? It may be using Integer or doing some sort of decimal rounding causing unintuitive behavior.

To Reproduce: Open attached file. Insert 'chr3:21,503,023-21,513,023' into "Quick Search". Click "Load Data". You should now see two reads. Right click on the "track title" choose "Color by..." choose "Score" and click "Edit". At the top of the heat-map gradient chooser you should see 4 triangles. Delete the 3rd one. Click the second one (YOU MUST CLICK IT OR TEXT ENTRY GOES NOWHERE!) Set the second one to 68.28 using the "Attribute Value" text entry. Hit return to set it (YOU MUST HIT RETURN TO SEE IT RENDER. This is not a problem here when we only have one threshold but when you have multiples you must always remember to hit return!). You should now have a screen like the screenshot [See attachment] Click "OK"

Observe: The threshold is 68.28 making anything above this white. Yet 68.29 is still black.

Conclusion: Behavior is unintuitive to an analyst that is purposely setting thresholds to make max scoring reads stick out.

Found on IGB 8.0.2 (16020)