Hi (Herve), I have been playing around with GappedAlignmentPairs and found it pretty useful. Some things that would be nice: 1) A method for getting the observed fragment length (would need to be NA when pairs are on different chromosomes). This should at least have an option (if not always) to exclude the N cigar runs from the calculation. Or it could just take it from the TLEN (isize) column in the BAM file (which at least in the case of gsnap excludes the Ns). 2) A method called "introns" or something that returns the N regions as a GRangesList, as a complement to grglist(). Such a method would also be nice on GappedAlignments. Thanks, Michael [[alternative HTML version deleted]]