This is probably misleading. For the sake of calculating base pairs, I completely agree that we should not consider nucleobases with highly distorted conformation. However, when counting how many nucleotides are there in the sequence (e.g., when generating ct or dbn files), nucleotides with distorted base conformation (or even nucleotides without their base groups) are still nucleotides. It may not be a good idea to just delete them from the sequence without at least printing out a warning message.