 So how well does this work? Well, it depends. You need to be lucky. So remember, on the one hand we had the actual robot, and on the other hand we had the computer where we're doing things with docking. It wasn't a coincidence that I listed those two compound screens projects before, because they had also been done with docking. In the first one, Lactamase, the docking actually returned two hits. Not only did it return two hits, but from the docking the result were two hits that were actually later experimentally confirmed. So there were two correct hits. And sorry, this said zero. Two is not a lot, but it's certainly a lot better than zero. And particularly when those zero came at the cost of $300,000. Second one, Curtain, when we had 146 experimentally, here the docking didn't do so well. We only found five. The reason for that is possibly chance, I don't know. But again, 146 is certainly better than five, but maybe these five contain slightly different groups, something to work with. At this point we're nowhere near a finished drug, but this is where we're going to need to take things back into the computer and lab and keep iterating and see if we can improve it. Why do we need to improve it? Well, because at this point it's not going to be very efficient. It might have a trace of an effect, but this would never make it even to the clinical trials.