Thanks for the encouragement for writing a paper on SNAP -- it's been on my to-do list for a while, but delayed for various reasons. Overall, I take it as a positive thing that a method paper is written after that corresponding program has been in active use for a while. My goal here is not to write a paper but to solve a set of related problems so that the community can build upon my work.
You are right in that pseudo-pairing/stacking interactions are between planar moieties in proteins and the standard base reference frame. The planar moieties include the amino-acids { "arg", "phe", "tyr", "trp", "his", "asn", "asp", "gln", "glu" } and the peptide bond. A reference frame is defined for each of them. The pseudo-pairing/stacking interactions of these planar moieties with nucleobases are identified and quantified using exactly the same algorithms as in 3DNA/DSSR. In addition to the pair-wise interactions, 'multiplets' and stacks (as in DSSR) involving both amino-acids and bases will be reported in future releases of SNAP.
HTH,
Xiang-Jun