Profiler for a function of a function

[#1]

I fitted a B-spline model (cubic, 5 knots) to the data from 50 pH curves (pH vs Volume) using functional data explorer. I then created a least squares model to predict the functional principal components based on chemical composition.

What I would really like to achieve is a profiler of predicted pH vs volume with sliders for chemical composition - predicting a pH curve for given composition. Because chemical composition is not directly in the formula for predicted pH, this doesn't seem to be possible. Wondering if there is any way around this?

[#2]

this might be relatively easy to do.  When you launch the Functional Data Explorer, add your chemical composition columns into the supplementary variables role in the launch dialog.  After cleanup and fitting, click on the Functional DOE Analysis option.  (you must have supplementary variables for this to work)