Apart from the comments about specifics in questions (which I share) and without knowing what it is you want to do there is one serious point here: the beauty of bioinformatics is: you (for the most part) don't need anybody's permission anymore to do anything. The data is out there, Google is your friend so go nuts. Just do it. Bioinformatics is one of the few disciplines where that is the case - there are very few gatekeepers (no PIs with labs you need to get into, no unis you need to be accepted into). You need a "torchlight" (i.e. your computer) and be willing to "pay for your batteries", but that's it. So go wild....:-)

You may look at the BioSQL project and its schema. Another example may be the complex Ensembl MySQL schema, which can be very difficult to understand.

ah, and this is the first biological database to have been built, to my knowledge: The Atlas of Protein Sequence and Structure

In any case, your question is too generic, and is difficult to answer...

You could impress people by showing off your cross-database querying of Bio2RDF-linked databases, using a non-faceted, perhaps natural language, approach.

Swan, one of your question tags states that the question is homework. I'm going to guess from this that you are a degree or masters student and perhaps not too familiar with asking questions in forum like this.

This is a great forum full of experienced professionals who are very willing to help even novices in the field. However you need to be much more specific in your question if you want to get any help.

What does the tool that you have written do? You say it integrates databases but to what end? You must have had a goal in mind? Unless you have written something so powerful it can interpret the schema of any biological database and re-write a new merged database coping with cross platform difficulties because that would be some homework!!

Please help us to help you.