Thanks, Kevin. I appreciate the acknowledgment and we're also excited to see the progress there. And let me take a shot at addressing your question. And if there's some gaps, Chris might be able to fill them in and you just let us know. But when you talk about the distribution across applications areas, discovery and translational versus the clinical research side, I would say that the needle, if you look at the global distribution, is definitely pointed more towards the clinical research side. And I would say it's probably 65% clinical research, 35% discovery and translational. And then within that clinical research segment, the applications are mostly around genetic diseases. And the portion that is focused on hematologic malignancies and solid tumors is increasing. We don't see a strong distinction there because they end up being -- meeting the 65-35, because they end up being pretty similar sites and labs. And whether you're running a steady supply of samples coming in for analysis based on an order from a physician or whether they're running samples for a clinical trial or a large research study, they all tend to be comparable in volumes. And so when we go back and we ask ourselves. Okay, what's the average? The average pull-through you can calculate, right? You have the installed base, it was 141. You have the consumables revenues, that's in the queue. And so given that you know that, that's why you've asked me to tell you what's the sort of like significant pull-through representative group. And what I would say -- the way I would answer the question is that significant fraction, 65%, 70% of adoption is through the reagent rental vehicle for us. And just to remind you that, that standard contract is $132,000 of overall consumables revenues, chips, reagents and everything, on an annualized basis. So if we think that 65-or-so percent of the systems that are going out in the last couple of quarters are representative of that pull-through, I think that that's how I would kind of start to factor toward that higher-end pull-through number. And there are sites that exceed that, right? You have sites that go beyond that and their ordering -- exceeds that. But so far, that's -- we're kind of adhering to that contracted volume, which is probably why we picked it in the first place. It's a good number and it fits the labs that are adopting the technology.