We thought about this possibility:
These sequences are self-complementary (i. e., palindromic). Therefore, for the sake of avoiding artifacts when having both 5' ends biotinylated we think it might be suitable to wash with 6M urea to turn the double helixes into their separate strands and remove every strand except those immobilized to the surface (we think that the biotin-streptavidine interaction can stand this treatment). The next step would be to add a saturating concentration of non-biotinylated DNA combined with heating to get the duplexes back, so they are biotinylated and fixed only by one of the 5' ends.
Do you think this is feasible?
Thanks again!
Luis