SeqPup should be free of this problem, but it will take
some help from users such as yourself willing to test it
to say for sure.
I've had a prior report of something similar to what you describe
but much less data lost (only part of the last sequence in a file of
several). The user avoided this bug by ending the file/document
with a dummy/junk sequence entry.
I don't know what is causing it. More importantly,
SeqApp won't be upgraded or fixed, as SeqPup is taking its place.
So if you have the interest in seeing this program mature and
get its bugs squashed, please test SeqPup(*) and let me know where
it is failing.
* from ftp://iubio.bio.indiana.edu/molbio/seqpup/
-- d.gilbert--biocomputing--indiana u--bloomington--gilbertd at bio.indiana.edu