DNA-level constraints on sequence evolution

Unless I have missed something, the criterion of RNA function used here is that an RNA sequence folds and similar folds may be found in the same RNA in related species. For example, the folds in ribosomal RNA may be conserved between species and ribosomal RNA functions as part of the ribosome by virtue of its folded structure. But until the "fRNAs" noted here are shown to have true function, we should not exclude the possibility that the folding serves a primary purpose at the DNA level, so that the RNA transcribed from that DNA has the folded structure by default. Indeed, a DNA sequence can be seen as a compromise between the competing demands of (i) DNA function, (ii) RNA function (if the RNA transcribed from the DNA has an independent function), and (iii) protein function (when the RNA is mRNA). Studies of this in my laboratory in recent decades have been summarized in a textbook (2006 "Evolutionary Bioinformatics" Springer, New York; see: http://post.queensu.ca/~f... ). For a recent discussion of the conceptual issues, please see the Journal of Theoretical Biology (248, 745-753; http://post.queensu.ca/~f... ).