thanks for all who replied to my original posting. The bottom line appears to
be that various groups have been using a "user-driven" approach to iterative
alignment - model optimization, but no generally recommended or automated
protocol appears to be in use.
thanks once more,
margot