Pacific Biosciences of California, Inc. (PACB) Q2 2014 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the Pacific Biosciences of California, Inc., Second Quarter 2014 Earnings Conference Call. [Operator Instructions] As a reminder, today's conference is being recorded. I would now like to introduce your host for today's conference, Ms. Trevin Rard. Ma'am, you may begin.

Thank you. Good afternoon, and welcome to the Pacific Biosciences Second Quarter 2014 Conference Call. Earlier today, we issued a press release outlining the financial results we'll be discussing on today's call, a copy of which is available on the Investors section of our website at www.pacb.com, or alternatively, as furnished on the Form 8-K available on the Securities and Exchange Commission website at www.sec.gov. With me today are Mike Hunkapiller, our Chairman and Chief Executive Officer; Susan Barnes, our Chief Financial Officer; and Ben Gong, our Vice President of Finance and Treasurer. Before we begin, I'd like to remind you that on today's call, we'll be making forward-looking statements, including plans and expectations relating to our financial projections and products that are subject to assumptions, risks and uncertainties and may differ materially from actual results. These risks and uncertainties are more fully described in our Securities and Exchange Commission filings, including our most recently filed quarterly report on Form 10-Q. Pacific Biosciences undertakes no obligation to update forward-looking statements. In addition, please note that today's call is being recorded and will be available for audio replay on the Investors section of our website shortly after the call. Investors electing to use the audio replay are cautioned that forward-looking statements made on today's call may differ or change materially after the completion of the live call. And now let's turn the call over to Mike.

Thanks, Trevin. Good afternoon, and thank you for joining us today. We are pleased with the second quarter 2014 results and the progress we are making in driving our overall business. Highlights of our second quarter financial results are as follows: We delivered 8 systems during the quarter, and with over 100 systems in our installed base now, we are becoming a well-established player in the sequencing space. Instrument revenue for the second quarter was $4.7 million, up 73% from the prior Q2 estimate revenue of $2.7 million. Consumable revenue for the second quarter was $3.0 million, up 62% from the prior year Q2 consumable revenue of $1.9 million. This is being driven by strong utilization growth on an increasing installed base. We're now averaging over $120,000 in consumable revenue per year per installed system. Total revenue for the second quarter was $11.4 million, up 89% from $6 million in Q2 2013. Year-to-date, our revenues for the first half of 2014 are up 98% from our first half 2013 revenue. On that note, it is clear from a revenue trend that PacBio is continuing to gain commercial traction. Our bookings in Q2 of 5 units were lower than the number in the previous quarter due to some customer logistical and tender delays that can cause our quarterly orders to fluctuate, but our pipeline continues to grow. With our installed base growing and system utilization increasing, our business has never been in better shape. We continue to forecast 70% growth in our revenue this year compared to 2013. Now I would like to provide some highlights from industry conferences we attended since our previous earnings call. At this year's American Society of Microbiology meeting in Boston, there was a very interesting presentation by Dr. Julie Segre from the National Human Genome Research Institute at NIH, on hospital-acquired infections. Dr. Segre described a study of patients from the NIH clinical center that were infected by an antibiotic-resistant bacterium called Klebsiella, which she referred to as "the nightmare bug." A total of 18 patients in the hospital were infected by the bacterium in 2011, but it was not clear at the time whether these patients had infected each other or had acquired an infection independently. With this study, Dr. Segre and her team decided that whole genome sequencing could help them build a phylogeny to recreate the history of these Klebsiella cases and determine how transmission was occurring. The project was completed using the PacBio platform because they needed to have high-quality genomic and plasma sequences for the best view of the microbes, which can get their drug resistance from a variety of Plasmids. The team studied samples taken from various patients and used epidemiology skills to build a transition map of the outbreak. Based on genome sequence and Plasmid diversity, the team determined that there were 2 separate infection events. Some transmission cases from the first patient and an independent infection from another patient. They also found horizontal gene transfer from patient to environment, highlighting a path that could explain transmission within the hospital. In her talk, Dr. Segre noted that once sequencing technologies such as PacBio are ready for clinical use, it will be very useful in stopping this kind of outbreak because they can identify clinical infection elements much faster than existing technologies. In June, we attended the 28th Annual European Immunogenetics and Histocompatibility Conference in Stockholm. The European Federation of Immunogenetics, or EFI, hosts this meeting each year, bringing together delegates from around the world to discuss basic and clinical science topics in the area of immunogenetics. At this is year's conference, attended by over 1,100 delegates, HistoGenetics and Anthony Nolan, who are both world leaders in HLA typing, presented their planned approaches for using SMRT sequencing to improve the quality of their HLA typing services. As a reminder, HistoGenetics and Anthony Nolan have both recently purchased multiple PacBio RS II systems, and they are in the process of integrating them into their mainstream service businesses. Dr. Nezih Cereb, CEO and Cofounder of HistoGenetics, described how they have evolved in their use of sequencing technologies for registry typing. HistoGenetics was the first HLA typing service to adopt Sanger sequencing. And until recently, it was the primary method of producing sequence-based typing information. Last year, they incorporated sequence data from aluminum MiSeq systems, and now they're adding PacBio to the workflow to improve the accuracy of their results. This presentation of Dr. Cereb gave examples of how tandem repeat structures in HLA regions were misaligned with MiSeq data. With the full-length sequences from PacBio, these same regions were aligned correctly. For HLA typing, it is critical to find accurate matches for transplant candidates, and therefore, obtaining the highest-quality information for HLA alleles in the registries is paramount. Neema Mayor, from Anthony Nolan Research Institute presented results of one of the experiments performed to validate the use of PacBio for their registry database. This experiment's 7 samples from the U.K., containing commonly seen HLA Class I alleles were sequenced. A total of 42 alleles were identified in these 7 samples, 37 of which were previously characterized and accurately matched the database sequences. The 38th allele first appeared to be an error when compared to the reference but was later validated as correct by cloning a Sanger sequencing, leading to a correction in the registry database. The other 4 discrepant alleles were also validated by cloning and Sanger sequencing and added to the latest build of the IMGT HLA database as new allelic variants by virtue of valid differences in the intronic regions of these genes. This experiment has led Anthony Nolan to conclude that SMRT DNA sequencing enables definitive HLA allele typing to single-sequencing reaction, and allows for simultaneous genotyping as well as new allele discovery. They believe this will ultimately have a considerable impact on matching potential donors and recipients. This past quarter, we also had 2 user group meetings, one in Asia and another on the East Coast U.S. We held our Asian group meeting on the heels of the Plant and Animal Genome, or PAG, Asia Conference in Singapore. At the PAG conference in our user group meeting, a research team at Temasek Life Sciences Laboratory in Singapore presented very impressive data on their sequence of the sea bass genome. As background, the Asian sea bass is an important fish for aquaculture, and therefore, there is great interest in better understanding its genome. The sea bass genome has approximately 700 million bases, or roughly 1/4 of the size of a human genome. Previously, the group at Temasek tried to sequence the sea bass genome with the Illumina HiSeq, augmented with some Sanger sequencing, but the results were not satisfactory to them. So they decided to sequence the genome with PacBio and obtained the following results: the total assembly size was 671 million bases compared to 622 million bases for the Illumina assembly. This suggests that at least roughly 7% of the genome was missing from the Illumina assembly. The number of contigs or pieces after assembly was 5,000 with PacBio compared to 122,000 with Illumina. This is a 24x improvement. The contig N50, which is the proxy for the average length of each contig, was 1.1 million base pairs for the PacBio assembly compared to just 32,000 for the Illumina assembly, which translates to a 33x improvement. The Temasek team is delighted with the high-quality results obtained using PacBio, and they are continuing to refine their assembly. Our East Coast user group meeting is an annual event that has been growing significantly. This year's meeting was held at the University of Maryland, and attendance was up 25% over last year with over 100 participants from various customer sites. There were 9 customer presentations with applications topics ranging from antibiotic-resistant bacteria, expanding reference databases in microbial area, the structural variant analysis and whole genome assembly in the human space. In addition, there were multiple presentations on software tools being developed for speeding up the assembly of large genomes. We also hosted full day bioinformatics training sessions for users interested in taking full advantage of the latest tools available. These training sessions are very popular among our user base with a total of 66 people attending the bioinformatics sessions. Providing efficient, easy-to-use software tools is a significant priority for us to drive further adoption of SMRT sequencing. In a recap of our user group meeting, the trade channel Bio-IT World reported on their observations of PacBio's commercial progress, concluding that, "The ability to sequence genomes and fragments thousands of base periods long has opened up projects that would be monumentally difficult and thought impossible with standard next generation sequencing." Shifting now to publications, we are continuing to see a growing number of customer publications citing SMRT sequencing at a pace that's making it difficult to keep count. One publication appearing in the Proceedings of the National Academy of Sciences and worth highlighting came from Dr. Michael Snyder's group at Stanford, entitled Defining a Personal, Allele-specific, and Single-molecule Long-read Transcriptome. As background for the study, the authors described how RNA sequencing with short-read technology is fairly common, but there are significant drawbacks. They go on to say, "Recent work has shown that reconstruction and qualification of transcript isoforms from short-read resequencing is insufficiently accurate." In order to attain what the authors described as a personal transcriptome in which all of the individual's genetic variants and transcript isoforms are defined and quantified for full-length transcripts, they sequenced the transcriptomes of 3 family members -- a child and both parents -- by using a PacBio long-read approach complemented with Illumina's short reads. The PacBio data enabled the researchers to characterize previously unidentified isoforms and to connect variants to RNA haplotypes. By assigning full-length RNA molecules to their transcribed allele, they could determine from which parent the molecule was inherited. Overall, the authors state that, "These results described, to our knowledge, the first large-scale and full-length personal transcriptome." Isoform sequencing with PacBio is still in its infancy, but we're very optimistic about the value it brings as evidenced by the results described in this and other publications. In summary, we are excited to see continuing momentum in several areas. Microbial area is growing in excitement around the study of antibiotic-resistant bacteria. This is driving more interest and growth in SMRT sequencing from government agencies, such as the NIH and the CDC. The plant and animal area, we're seeing more de novo sequencing projects as researches have taken note of the success that early customers have had in creating high-quality reference genomes for plants such as spinach and melon. These are projects that have historically been taken on using Sanger sequencing because short-read sequencing technologies have been less effective in creating useful references for these large, complex genomes. With our advances and throughput over the past 2 years, researchers are finding that it is now more economical to approach these projects with PacBio. Many of our core lab and service provider customers have taken on large projects, such as sequencing the rice and cotton genomes. In the human sequencing space, there's growing interest in sequencing clinically relevant genes containing structural variation that is too delicate to sequence with short-read technology. In addition, there is significant momentum building around our solution to HLA typing. Last month, we announced a co-marketing arrangement with GenDx, who provides solutions for amplifying highly polymorphic HLA genes so they can sequence with the PacBio RS II. This is a great solution for customers who do not have their own assays for generating HLA samples. Overall, we are seeing great interest across multiple applications, which is driving growth in our consumables business and building a pipeline for new system sales. Lastly, I would like to note that our collaboration with Roche is going very well. We're expecting to complete our first project milestone with them later this year. This is consistent with the schedule we have set out with Roche when we entered into our agreement in September of last year. That concludes my remarks, and I'll now turn the call over to Susan.

Thank you, Mike. And good afternoon, everyone. I will begin my remarks today with a financial overview of our second quarter that ended June 30, 2014. I will then provide details on our operating results for both the quarter and year-to-date with a comparison to the second quarter of 2013 and year-to-date comparison to the first half of 2013, respectively. I will conclude my remarks with a brief discussion of our balance sheet. Starting with our second quarter and year-to-date financial highlights. During the second quarter, we recognized revenue of $11.4 million, and incurred a net loss of $19.1 million. This brings our year-to-date total revenue to $23.1 million and our net loss to $38 million. We ended the quarter with $105 million in cash and investments, $13.7 million lower than the $118.7 million reported at the end of Q1, and $7.5 million lower than the 2013 year-end balance of $112.5 million. As a reminder, in the first quarter this year, we raised $20.6 million in equity funding from our aftermarket facility. We did not issue any equity under this ATM during the second quarter. Excluding these proceeds, cash use totaled $28.1 million year-to-date. Turning to revenue. Revenue increased substantially in Q2 year-over-year across all products and services. Total revenue for the quarter of $11.4 million was 89% greater than the $6 million recognized in Q2 of 2013. Year-to-date revenue in 2014 was $23.1 million, up 98% over revenue of $11.6 million recognized in the first half of 2013. Instrument revenue increased 30 -- 73% to $4.7 million in Q2 2014, from $2.7 million during the same period last year. Year-to-date, instrument revenue increased 117% to $10 million in 2014 versus $4.6 million for the first half of 2013. The increase in revenue reflects 8 instruments recognized during the second quarter and 17 year-to-date in 2014 compared to 3 instruments recognized in Q2 2013 and 6 total for the first half of the year in 2013. Consumer revenue was also strong during the quarter, increasing 62% to $3 million for the current quarter, up from $1.9 million reported during the second quarter of 2013. Year-to-date, consumable revenue increased 46% to $5.6 million in 2014 compared to $3.8 million in the first half of 2013. Service and other revenue increased 37% to $2 million in the quarter compared to $1.4 million in Q2 of 2013 and up 27% year-to-date to $4.1 million in 2014 from $3.2 million in 2013. And finally, we continue to recognize $1.7 million of revenue per quarter associated with the $35 million upfront payment that we received from Roche in Q3 2013. Gross profit was $3.1 million in Q2 of 2014, representing a gross margin of 27% and year-to-date, gross margin recognized in 2014 was $5.8 million, corresponding to a gross margin of 25%. In Q2 of 2013, we recognized $1.1 million of gross profit with a gross margin of 18% and year-to-date gross profit in 2013 was $2 million with a gross margin of 17%. As a reminder, gross profit in 2014 includes $1.7 million of margin per quarter relating to the Roche revenue. The increase in gross margin in 2014 has been tempered by a change in product mix. We recorded a larger number of instrument revenue for both Q1 and Q2 2014 in comparison to the first half of 2013. Since instrument revenue has lower margins than our consumables, the mix of instruments to consumables affects the gross margin percentage result. Our revenues for the quarter and the year were up significantly year-over-year, operating expenses were relatively flat. Operating expenses in the second quarter of 2014 totaled $21.4 million compared to $21.1 million incurred in Q2 of 2013. Year-to-date, total operating expenses were $42.3 million compared to $42.6 million in the first half of 2013. Breaking down our operating expenses. R&D expenses in the quarter were $12.4 million, up slightly from $11.7 million of expenses in Q2 of 2013. Year-to-date, expenses in R&D were $24.2 million, also up from the first half of 2013 expenses of $23.7 million. In both comparisons, R&D expense increases were primarily a result of higher product development expenses incurred in Q -- in 2014 versus 2013. R&D expenses this quarter included $800,000 of noncash stock-based compensation expense. SG&A expenses for the quarter were down in Q2 2014. We had $9 million in SG&A expenses compared to $9.4 million incurred in Q2 of 2013. We have incurred $18.1 million so far this year compared to $18.9 million incurred in the first half of 2013. SG&A expenses in 2014 have come down across many categories, as we continue to pursue SG&A related cost efficiencies. SG&A expense for this quarter included $1.3 million of noncash, stock-based compensation expense. Also, in the area of other income and expense, in Q2, we reported $800,000 of other expense, primarily related to the interest expense associated with debt that we took on in Q2 of 2013. Year-to-date, we have incurred $1.5 million of other expense. Ben will provide further guidance on our ongoing expense rate later in the call. Now turning to our balance sheet, as I mentioned at the beginning of my comments, cash and investments decreased $13.7 million to $105 million at the end of the second quarter, primarily reflecting our second quarter net loss of $19.1 million. That's $3.3 million of noncash expenses from stock-based compensation expense and depreciation. Balance sheet changes also contributed to lower cash use in the quarter. Inventory balances rose $1.1 million in Q2 to $9.4 million from $8.3 million at the end of Q1. This increase is necessary to ensure the timely fulfillment in product orders being generated in 2014. Accounts receivable increased to $4.3 million at the end of Q2 compared to $3.4 million at the end of Q1. This concludes my remarks on the financial results for the quarter. I would like to turn the call over to Ben.

Thank you, Susan. I will be providing an update to our 2014 financial forecast. Starting with revenue, we are pleased with the continued growth we saw in Q2, and we are maintaining our forecast of 70% growth in revenue for the year. This translates to a total annual revenue estimate of $47 million to $48 million for 2014. With regard to the third quarter, we had 10 instruments in backlog at the end of Q2, and we expect to deliver a majority of those systems in Q3. Moving on to gross margin, as we have seen in the past 3 quarters, $1.7 million quarterly amortization of the Roche payment from last September continues to result in favorable year-over-year gross margin comparisons. Our gross margin at 27% for the second quarter was in line with our expectations, and our forecast for the year remains similar to what we communicated last quarter with a slightly tighter range. We are forecasting our gross margins to be between 25% and 27% for the year. As a continuing reminder, at our current revenue levels, small changes in gross profit dollars can cause fluctuations in our quarterly gross margin percentage. Our operating expense for Q2 at $21.4 million was a little higher than the previous few quarters due to slightly higher research and development expenses. We plan on maintaining our quarterly operating expense at around this level for the rest of the year. Our operating expenses include noncash stock compensation expense and depreciation expense that together amounts to between $3 million and $4 million per quarter. Net interest expense of $800,000 for the second quarter was a little higher than our previous estimate due to a revaluation of the noncash derivative component of our debt. Excluding the revaluation, which amounted to $150,000, our net interest expense was in line with our previous estimate. We expect to continue to record approximately $650,000 in net interest expense per quarter this year. With regard to cash use. $14 million this past quarter and $28 million year-to-date, excluding ATM proceeds, puts us right on target to meet or beat our previous forecast of $60 million for the year. Taking into account the $21 million in additional cash we raised in Q1 from the ATM, we expect to use less than $40 million for the year. Therefore, we should end the year with at least $73 million in cash and investments. Now with that, we'll open the call to your questions.

[Operator Instructions] And our first question comes from the line of Bryan Brokmeier of Maxim group.

You started off talking about the bookings and the decline sequentially. I was wondering if you could provide us a little bit more color. And also, in terms of the different conferences that you mentioned that you attended during the quarter, were any of those bigger drivers of sort of your pipeline for future bookings? And what sort of research those customers are doing within your pipeline?

Okay, if I could fill a lot of that. The issue with orders was driven by a few cases, as we tried to point out, of sort of customer logistical and tender delays, which took longer for them to get through the process than we and they would have hoped. So we see that as, hopefully, a picker [ph] blip in that. It wasn't a lack of order prospects in our pipeline at all. That actually has continued to increase. But given that -- and we tried to emphasize this before, we're still a relatively small total order number per quarter, and a couple of those sites having a logistical issue, either because their tender process is slow or because they're having to wait on some other event, a lab being secured and so forth, can cause quarter-to-quarter blips in that. But it didn't concern us greatly. Relative to the conferences, I think we certainly saw a really healthy interest in the HLA community. This was really the first time that we had, had much of a public presence at an international conference at this level in the immunogenetics space. It was the first time that our early customers at 2 of the largest typing centers in the world had, had a chance to present their experience with the PacBio RS II system. We even had some relatively bluebird presentations by some of our other customers in the academic world who were doing substantial HLA typing that we didn't know about, really, until the presentation, till the conference. So we had a lot of interest from a lot of different kinds of laboratories interested in the whole HLA space, not just in the major typing centers, but in a lot of the other labs that may be engaged with pharmaceutical companies, who are very interested in having high-resolution HLA typing in done conjunction with clinical trials and other large genomic projects. So I think we rated that conference as a real success from a prospect perspective. In the Asia conference, I think it was a continuation of the interest we're seeing within the plant and animal world to kind of reengage with genomic sequencing in order to come up with much better reference genomes for a whole variety of organisms than they've been able to successfully generate with the combination of short-read and all-Sanger methodologies, which are both slow and expensive but also just don't give you the kind of results that they really want and need within that space. But I think we saw both of those conferences as very successful marketing opportunities for us.

Have you seen some of those opportunities in your pipeline turn into bookings since the end of the quarter?

Well, we're not talking about the results for this quarter, but we certainly have seen a fair interest of those turning into sort of proof-of-principal type runs with people particularly in the relatively new -- and the HLA space. Within the plant and animal ones, we have an ongoing buildup on our pipeline there.

Then, just the last question...

[indiscernible] across the world.

Okay. And just lastly, on your Roche collaboration. You said you expect to achieve the first milestone later this year. Any more color on what that milestone is? Is it completing an instrument? Or any details about what you've accomplished so far?

No. What we said and what Roche has said publicly from the beginning, is that we assume that getting a diagnostic level device created would be at least a 3-year project, and we set out with them at the beginning a series of milestones as part of that development process, but we've been unwilling, for competitive reasons, to go through the details of what those milestones were going to be and what the specifications associated with that system were, and I think we're going to stick with that.

And our next question comes from the line of Bill Quirk of Piper Jaffray.

First question from me is just, I want to talk a little bit about consumables. Consumers bounced around a little bit up in the fourth, kind of off a little bit in the first and then back up here in the second. And so, just trying to get a little better sense, I guess from your perspective, kind of how do you see this bouncing around? Obviously, it's moving generally in the right direction. Are we at a level now where we consider this to be sustainable with upside to come? Just help us think a little bit about this.

Well, it's like anything else, they bounce around a little bit. We're still a relatively small total number so that a few big orders can change 1 quarter. I mean, some people will order on a 6-month basis, some on a 6-day basis. So it does matter when you get big orders in chunks. I would say that we're quite happy with the overall increase. And If you kind of step back and look at it, it tends to -- in some cases, it's gone up and, then it's leveled off 1 quarter, and it's gone back up. Draw a line through more than 2 data points, you get a pretty good trend, I think. And that's kind of where we're gauging our ability to forecast that. [indiscernible] uptick this year and sort of across the world, I think, and we're seeing, as we kind of pointed out, an increase in per customer usage. I think the last guidance we'd given you was somewhere between 100 to 120. And we're now saying it's north of 120, on average, for the year.

And Mike, just kind of going back to the earlier question around bookings, there wasn't any effect from any of that competitive dynamics in the market. Your still -- your funnel, your sales funnel, seems to be in pretty good shape. Is that the message you're trying to tell us? No issues around competition?

Well, I wouldn't say there's no issues around competition, but I don't think that was a substantial issue for us in the quarter. I mean, we compete in a select set of applications in the marketplace very well. We're not a head-to-head competitor with any of the short-read technologies, either Illumina or Ion Torrent, for most things. We tend to focus on those areas and our customers do, where those technologies don't perform so well at this point. So that's where we go. And I don't think the competitive landscape in that space has really changed very much.

Okay. Great. And then just lastly for me. You typically highlighted on the call but just any update with respect on the planned system upgrades?

Upgrades, yes, so we do plan on putting out more system enhancements during the course of this year, not hardware ones, but we actually kind of lined this out in our investor presentation earlier where at least a couple of times a year, we put out software enhancements and we put one out, I think around April this year, so we're kind of on schedule to do another one in the latter part of this year. And we also typically put out some sort of chemistry enhancement, and we're also on target to put one of those out later on this year. We haven't given specifics as to what the improvements are going to be on those -- from that release, but it's not too far in the future. So you should keep an eye out for that.

Yes, what we have talked about is, generally, that they will represent an increase in average read length and throughput, improvements in sample prep that you have, again, high utilization of the system. And we will be rolling out, actually, a series of those in the last half of the year.

And our next question comes from the line of Amanda Murphy of William Blair.

It's actually J.P. in for Amanda. I just had a couple of questions around consumables usage. Do you guys see kind of a large chunk of customers running at full capacity and maybe others not so much? And then can you just remind what me what the theoretical max pull through per just one machine is?

I'll take a first shot at that. So it's not an 80-20, meaning, it's not so concentrated that 20% of our customers are doing 80% of the consumables. It's actually more spread out than that. I think with any capital, you might expect that there's going to be some customers that are not as productive as others. We certainly have that. But I'd have to say that the overall usage is spread across a large number of users. And to answer the question on the high end, we definitely see multiple customers who've been able to do a run rate of over $300,000 per year. So we think there's adequate run rate in terms of that increase and average pull-through. As Mike said, it's not greater than $120,000, but that means there's still a lot of capacity left out there for improvement.

And then just one more. Can you provide additional detail on the size of the HLA market? I mean, have you guys looked at kind of a total addressable market there?

Well, yes, but it's -- let me give a caveat at the end of that. So I think the general estimates are that the available market in petitional HLA registry typing market is in the order of $200 million to $300 million a year worldwide. However, I think that the larger potential market there is actually not in a traditional registry type work but in association with clinical trials, being able to do HLA typing at high resolution for big trial projects. And we've seen a fair amount of interest from the pharmaceutical world as have several of the labs that traditionally do typing for registry purposes. There, as soon as the technology is appropriate to handle that high resolution, high throughput requirement that they have. And that's potentially from my perspective, maybe several times bigger than the traditional market, which is not trivial for us. I think it's been untapped because just the technology has not been there.

And then, are you guys able to provide a -- kind of a bookings outlook for the remainder of the year? Like, should we think about the logistical delays as just pushing into the next quarter? Or is it just going to push things down the line in general?

Yes, J.P., the -- we have not traditionally given a specific on bookings outlook, but the -- embedded in our revenue outlook of 70% growth, it suggests that there's going to be higher bookings than what we just saw in the second quarter because in order to achieve those revenues, we, obviously, need to have systems to install. And even with the growth we have in consumables, we'd probably need to continue to see some pretty healthy bookings in order to get to the 70% revenue growth. But that's on purpose that we've maintained that revenue forecast.

[Operator Instructions] And our next question comes from the line of Tycho Peterson of JPMorgan. Tejas Savant - JP Morgan Chase & Co, Research Division: It's Tejas on for Tycho. Can you share some color on the backlog in terms of new versus existing customers?

Yes, so the backlog of the 10 systems, we definitely have, in that backlog, additional systems for existing customers. And in fact, one of the bookings we did have this past quarter was also for an existing customer. So related to a previous question about utilization, when people get to a certain utilization, they do look at buying additional systems and then we had mentioned before that we've had multiple orders in the past and, usually, it takes a little bit of a while to get everybody installed on those multiple orders. Tejas Savant - JP Morgan Chase & Co, Research Division: And then, you guys have highlighted the ag opportunity and sequencing cash crops. Has that translated into any orders yet? Or is it still too early for that?

Yes, several of our existing customers are doing that. And in Europe, as an example, KeyGene is one of the premier sort of genomics/agricultural companies. They're one of our highest-capacity users on our systems as an example. But a lot of -- we have a system at the USDA, which is focused on plant...

Or labs.

And animal ones. A lot of the core labs do projects that are in that. Several of the service providers and our Asia customers are doing large plant and animal, mostly plant genome. So that's already a good part of our business. And as people get out the results of some of the bigger projects they've done in comparison to their other attempts with other technologies with shorter read capacity, then that's making a lot more people aware of the advantages of the PacBio platform. Tejas Savant - JP Morgan Chase & Co, Research Division: Okay. And now just quickly turning to Roche. Does your guidance for this year of $47 million to $48 million include the milestone payment? Or will that be separate?

That's separate. So the $47 million to $48 million is based off of just the same kind of revenues that are in the revenues today from Q1, Q2. Tejas Savant - JP Morgan Chase & Co, Research Division: Okay. And in terms of...

[indiscernible] what the size of that payment might be, either. Tejas Savant - JP Morgan Chase & Co, Research Division: All right. Any thoughts on their acquisition of Genia in terms of the nanopore sequencing effort announced in Roche and how that changes things for you, if it does at all?

I think we commented on that before. From my perspective, it doesn't -- and from what we understand from Roche, it doesn't change our relationship with Roche at all. They view this as another long-term opportunity for them to potentially come up with a technology that they could sell in more areas than the diagnostic arena, which is where our agreement with them is limited to. But they also recognize it's a very, very early technology with a high risk associated with it. So to them, it's another bed in the space, but one that they realize is fairly far from commercializable.

And our next question comes from the line of Zarak Khurshid of Wedbush Securities.

Zarak Khurshid at Wedbush. Can you tell us what the gross margins are on the consumables, currently? And taking it a step further, what revenue run rate is required for you to break into the kind of 50% gross margin range?

Zarak, this is Ben. We don't give the breakout of the gross margins, but we do let people know that from a mix perspective, the consumable margins are definitely higher than they are for the systems and the service. And sometimes you're subject to just the overall revenue being not high enough to cover amount of fixed costs. So I think your question has more to do with at what revenue levels can you get to perhaps breakeven. And what we've said in the past is somewhere between $100 million and $150 million in revenues on the top line, you'll probably get to breakeven. And from that, you could probably start to think about what margin that means to cover all the operating expenses.

And I'm showing no further questions at this time. I would now like to turn the call back over to Mr. Mike Hunkapiller for any closing remarks.

Okay. So in closing, we remain steadfast in our commitment to bringing the unique advantages of our SMRT technology and products to our customers and the scientific community in general. We believe that SMRT sequencing provides the industry's most complete and accurate picture of genomes due to its superior performance in sequencing accuracy, uniformity of coverage, extremely long read lengths and the ability to characterize DNA-based modifications. We continue to make progress in driving the adoption of our products and is rewarding to see momentum building in our business. We look forward to talking again in 3 months time.