Let's try to sort this out. What is on the disc is 8-bit 4:2:0 (DVD, BD, HD-DVD, it doesn't matter). Upsampling the bit-depth does not really get you anything until you start trying to do processing on the image stream. At the point at which you are doing most anything to an integer-based data stream, you want additional bits to mitigate rounding errors that might happen during whatever processing operation is going on. This is especially true for small changes (where the results often round to 1.0) and really big changes.
What is the net result: if you are leaving the controls in your player alone (we'll cover this in a second), then do not worry about "downsampling" on a 10-bit 4:4:4 signal. There is really nothing much there in the incremental bits unless something "funny" is happening in the player.
To calibrate with a VP in the chain, use the test patterns from the VP to calibrate the display's controls. EDGE includes a lot of fairly useful patterns (more than some affordable dedicated signal generators) to do this (e.g., grayscale, color bars, etc.). The goal is to make the display neutral with respect to the EDGE.
Once the link between the VP and the display is calibrated, then the goal is to neutralize the source component with respect to the EDGE. Using the controls on the EDGE or the individual source component, you want to calibrate the inputs for the individual source components to produce a fully-calibrated image on-screen. In other words, if you are displaying an appropriate color bars pattern from a test disc, and you need to change saturation or hue, then change this on either EDGE or the individual component, not on the display itself. Use the display's controls as a last resort.
Step 1: calibrate the display using test patterns from EDGE to make this link "neutral" with respect to color.
Step 2: calibrate each of the individual sources using thei own controls or the controls on the EDGE to make each of these links "neutral" with respect to color.
Important note: if you have a CMS in your display, then calibrate the gamut first, before you do grayscale, color decoding, etc. This will save you from having to chase your tail in some key areas, while also minimizing the amount of signal processing your display is having to do. This is contrary to what a popular thread in the display calibration forum says, but it is what we advise our users who should mostly ignore that thread.
Hope this helps!