Aside from the motif, you could also look at general homology to proteins with known N-linked glycosylation sites. If it were a different type of PTM, I might suggest looking at intrinsic disorder or a structural motif, but there doesn't seem to be a preference for structure or disorder in N-linked glycosylation. My only other thought would be to look at common interaction partners. Not just the oligosaccharyltransferases, but maybe some other proteins that provide some fine tuning? Sounds like a fun project!