There are a number of parts to that question.
I don't think the data bank alone does all of the suggestions you've made. One of the reasons I mentioned the U.K. was that they do have a longer history, and obviously a larger database. That's why the experience and statistics coming out from that country are helpful to us.
Another thing to keep in mind is that as a database grows--and on a statistical basis, this is called the cold-hit paradox--the bigger the database becomes, the chances of false matches grow.
Very quickly, just to give you an analogy, I'm probably the only Rondinelli in this room. Let's say that Rondinelli is my DNA profile, and the database is comprised only of us. Let's say we're all murderers and sexual offenders in here. There's going to be a chance that my Rondinelli will not match with anyone in here in a smaller base. But let's take the phone book for the Ottawa area. If you flip through it, there's probably going to be a better chance that you're going to get a hit with Rondinelli, because there are probably some cousins I have out here that I don't know about. So expanding a database has a whole host of other issues on a statistical basis as well.
DNA definitely plays a role in the criminal justice system in exonerating suspects, but it's not from the data bank that this happens. Police surreptitiously take samples from suspects all the time, then they match them to the crime scene sample. If they don't come to a match, they're struck off their list. That happens, and it's useful, but it's not from the data bank.