The three primary megablast databases available on the public main
Galaxy instance are comprised of individual fragments/sequences of
different types from many species (not assembled genomes):
If you want to use megablast to map against specific assembled
then using a local or (better) cloud instance is recommended. In your
own instance, the individual genomes would be set up the way that the
'phiX174' is set up on main.
To get started, please see: http://getgalaxy.org
Does this address your question? If not, perhaps you could explain
what your goal is and we can try to offer help or confirm that this is
the best path?