Illumina, Inc. (ILMN) Q3 2012 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the Third Quarter 2012 Illumina Incorporated Earnings Conference Call. My name is Chanel, and I will be your operator for today. [Operator Instructions] As a reminder, this conference is being recorded for replay purposes. I would now like to turn the conference over to Mr. Marc Stapley, Senior Vice President and Chief Financial Officer.

Thanks, Chanel. Good afternoon, everyone, and welcome to our earnings call for the third quarter of 2012. Before we get started, I'd like to announce that Rebecca Chambers, who has just joined Illumina, is our new Senior Director of Investor Relations. Rebecca comes to us from Myriad Genetics, where she was Head of Investor Relations and Corporate Communications. We're very excited to have Rebecca joined the team. Becca?

Thank you, Marc. I'm very happy to have joined the team as well, and I look forward to working with our investors and analysts. During the call today, we will review the financial results released after the close of the market and offer commentary on our commercial activity. After which, we will host a question-and-answer session. If you have not had a chance to review the earnings release, it can be found in the Investor Relations section of our website at illumina.com. Participating for Illumina today will be Jay Flatley, President and Chief Executive Officer; Marc Stapley, Senior Vice President and Chief Financial Officer; and Christian Henry, Senior Vice President and General Manager of our Genomic Solutions business. This call is being recorded, and the audio portion will be archived in the Investors section of our website. It is our intent that all forward-looking statements regarding our expected financial results and commercial activity made during today's call will be protected under the Private Securities Litigation Reform Act of 1995. Forward-looking statements are subject to risks and uncertainties. Actual events or results may differ materially from those projected or discussed. All forward-looking statements are based upon current information available, and Illumina assumes no obligation to update these statements. To better understand the risks and uncertainties that could cause actual results to differ, we refer you to the documents that Illumina files with the Securities and Exchange Commission, including Illumina's most recent Forms 10-Q and 10-K. Before I turn the call over to Marc, I wanted to let you know that we will participate in the Piper Jaffray Annual Healthcare Conference on November 28. For those of you unable to attend this conference, we encourage you to listen to the webcast presentation, which will be available through the Investor Relations section of our website. With that, I will now turn the call back over to Marc.

Thanks, Rebecca. During my section of today's call, I will review our third quarter financial results and our guidance for the rest of fiscal 2012. I will then turn the call over to Jay put the provide an update on our commercial progress and the state of our business and markets. We're pleased to report that our positive momentum continues, as this quarter marks the fourth consecutive quarter of sequential growth in both revenue and non-GAAP EPS. In Q3, strong sales for HiSeq and an increase in delivery of whole-genome service contracts led to revenue of $286 million, a 2% sequential increase. Included in these results is revenue of approximately $1 million from BlueGnome, the acquisition of which was announced on September 19. Year-over-year, revenue increased 21% due to the strength of our sequencing business. Instrument revenue for the third quarter was $82 million, up 14% sequentially, due to an increase in sequencing instruments, and up 15% year-over-year. The year-over-year growth was primarily due to the impact of MiSeq, which began shipments in Q3 of last year. Consumable revenue for the quarter was $177 million, down 4% sequentially and up 22% compared to the third quarter of 2011. Consumable revenue represented 62% of total revenue in line with Q3 last year and down slightly sequentially. The sequential decline has 2 primary components: the effect of reduced usage during the summer vacation season and the fact that Q2 was biased somewhat positively by success in our efforts to reduce reagent back orders. Services and other revenue, which includes genotyping and sequencing services as well as instrument maintenance contracts, was $23 million for the quarter, an 8% increase sequentially and a 54% increase year-over-year. Importantly, we shipped more than twice as many whole-genome samples from our FastTrack Services lab as compared to the prior quarter, a record for our services business. Turning now to gross margin and operating expenses. I will highlight our adjusted non-GAAP results, which exclude non-cash stock compensation expense, headquarters relocation expense and other non-cash items. I encourage you to review the GAAP reconciliation of non-GAAP measures included in today's earnings release. Our adjusted gross margin for the third quarter was 70.5%. This compares to 70.9% last quarter and 68.9% in the third quarter of 2011. The sequential decline was primarily due to an increased mix of instrument revenue. Compared to the third quarter last year, the increase was driven by a higher mix of sequencing consumable revenue. Adjusted research and development expenses for the quarter were $46 million or 15.9% of revenue compared to 14.7% of revenue in the second quarter, with 17.4% of revenue in the third quarter of 2011. The sequential increase was primarily due to headcount additions with clinical trial expense to support the planned FDA submissions of MiSeq and our cyto products. Adjusted SG&A expenses were $56 million or 19.5% of revenue in the quarter. This is flat compared to the second quarter and down compared to 22.2% of revenue in the third quarter of 2011. Adjusted operating margins were 35.1% compared to 36.7% in the second quarter, again, primarily due to higher research and development expense, as well as the mix of instrument and consumable sales. On a year-over-year basis, operating margins expanded significantly due to a considerably higher revenue recorded this quarter. In the third quarter, we recognized approximately $4 million of adjusted other income, which contributed $0.02 to non-GAAP EPS. Our non-GAAP tax rate for the quarter was 33.9% compared to 34.4% in the third quarter of last year. The tax rate was higher than expected as a result of the delay in the passage of the U.S. R&D tax credit. Non-GAAP net income was $54 million for the quarter and non-GAAP EPS was $0.41. This compares to non-GAAP net income and EPS of $30 million and $0.22, respectively, in the third quarter of 2011. The reported GAAP net income of $30 million or $0.22 per diluted share in the third quarter of 2012 compared to $20 million or $0.15 per diluted share in the prior year period. The current period results include $4 million of ongoing expenses related to Roche's unsolicited tender offer, as well as a $19.5-million charge associated with the relocation of our headquarters, as we completed the exit from our previous facility. Relocation charges are not expected to be material going forward. During the quarter, we generated cash from operations of $51 million. Capital expenditures were just under $18 million, resulting in approximately $34 million of free cash flow. This amount is lower than the $68 million of free cash flow in the third quarter of last year due to changes in working capital associated with the timing of shipments. Sequentially, free cash flow was impacted by higher accounts receivable levels. DSO increased to 69 days compared to 61 days last quarter, also driven primarily by the timing of shipments. We ended the quarter with approximately $1.2 billion in cash and short-term investments after using $88 million of cash to acquire BlueGnome. During the quarter, approximately 569,000 shares were repurchased for $25 million under our previously announced repurchase program, leaving slightly less than $200 million of authorization remaining. We are pleased with our financial performance for the first 3 quarters of 2012. The impact of the academic and research funding environment from this year's results is largely known due to the continuing resolution being signed into law. With the fiscal cliff and presidential election on the horizon, we do remain cautious. As a result, we are projecting fiscal 2012 revenue in the range of $1.134 billion to $1.144 billion, which equates to fourth quarter revenue of approximately $295 million to $305 million. Fiscal 2012 non-GAAP earnings per fully diluted share are expected to be in the range of $1.54 to $1.59 or approximately $0.38 to $0.43 in the fourth quarter. Importantly, these projections include expected gross margins of about 70% for the full year; the impact of the R&D tax credit, whether passed or not; additional share repurchases; and BlueGnome revenue of approximately $5 million. BlueGnome's impact on EPS is expected to be neutral in the fourth quarter. At this point, I will turn the call over to Jay for some remarks on our commercial activity during the quarter before we begin the Q&A session.

Thanks, Marc. As previously mentioned, we're very pleased with our third quarter results, particularly in light of the macro conditions in our markets. The positive trends, that began in the fourth quarter of 2011, continued through the first 3 quarters of 2012. Despite third quarter reagent utilization being somewhat impacted by summer vacations and the clearing of back orders in Q2, we saw sequential revenue and EPS growth with solid operating margins. This quarter saw the introduction of several new products and service solutions, many of them clinically focused, including the recently completed acquisition of BlueGnome. Sequestration aside, we now believe that the NIH funding is set for the next 6 months, given the continuing resolution that's been signed into law by President Obama. Additionally, the NIH recently stated that it will conservatively fund existing grant awards at 90% of the prior year's level for the next 6 months, given the uncertainty of the final budget. This is consistent with how the NIH has dealt with continuing resolutions in the last 5 years. As everyone is aware, it's unclear whether sequestration will actually take effect on January 2, and if so, for how long it will remain in place. Our outlook for European and Asian funding is unchanged. As consistent with our communication on the second quarter call, funding in these regions appears to be stable. We're seeing a diversification in our customer base, reducing our overall exposure to government funding. We've accomplished this by focusing on translational genomics in clinical segments and by increasing our penetration into applied markets. As a result, we believe that approximately 30% of our revenue today comes from non-academic and non-government segments. Going forward, we're committed to furthering this diversification through our focus on developing these markets. I'd like to now turn to the specific results of the quarter. Q3 shipments to the Americas were much improved over last year. We saw U.S. shipments increase 19% year-over-year, and Canadian shipments increase 11%. European shipments were up 19% over the same period, while shipments to Asia continued to be strong with growth of 21%, including 79% growth in China. In the array business, total microarray revenue declined 3% sequentially, primarily due to strong Infinium sample shipments in Q2. However, shipments for iScan microarray instruments in Q3 were the highest since the third quarter of 2010. We recently launched the Infinium HumanCore Arrays, which include the HumanCore, HumanCore Exome and HumanCore custom arrays. This family of products was developed in collaboration with several leading research institutions in an effort to provide a tool to perform large-scale, economical genotyping studies, with the option to customize content for a specific cohort or population. Turning now to our sequencing business. Total sequencing revenue in Q3 grew 4% sequentially, attributed primarily to growing instrument revenue and genome sequencing services. Total sequencing consumable revenue grew 44% compared to Q3 of last year. While HiSeq utilization was down sequentially, this was partially offset by strong growth in the HiSeq and MiSeq install base. TruSeq and Nextera sample prep revenue year-to-date has more than doubled as compared to the same period as last year, demonstrating the rapid uptake of Illumina sample prep kits within our customer base and the overall importance of complete end-to-end solutions for sequencing. Average annualized consumable utilization on HiSeq decreased to $315,000 in Q3 from $338,000 in Q2. It is up 17% year-over-year from $270,000 in Q3 2011. We expect quarterly variations and pull-through. Due to seasonality in the timing of purchasing cycles, we encourage investors to use a multi-quarter rolling average as a method to model pull-through trends. Consumable price increase announced in April will be almost entirely reflected in the pull-through for this coming fourth quarter. In Q4, we will also begin to see the impact of the rapid run mode of the HiSeq 2500, which yields greater revenue per genome, and we expect sample prep revenue to continue to grow. In September, we announced the TruSeq Stranded mRNA and total RNA sequencing sample prep kits. These kits enable gene expression studies that provide a complete view of the transcriptome even from low-quality RNA samples, such as formalin-fixed paraffin-embedded or FFPE samples. Total RNA kit leverages Epicentre's Ribo-Zero ribosomal RNA reduction chemistry, an example of how we're combining unique technologies from Epicentre with the core Illumina sequencing technology to provide complete solutions. These new kits have been received positively by customers, who praised the high data quality, workflow improvements and ability to analyze FFPE samples. This feedback, particularly in view of the ENCODE project which highlighted the enormous potential for RNA-Seq, enforces our excitement about the future potential of this application. Sequencing instrument revenue was up 16% sequentially, primarily due to increased HiSeq shipments. We continue to be delighted by the interest in our highest throughput instrument, the HiSeq 2500, both from new customers as well as core labs and research centers that are adding capacity. We're also excited about the rate of upgrade orders for the HiSeq 2000 install base. Based on orders booked to date, at least 30% of the install base of HiSeq 2000 and 1000 instruments will be upgraded exceeding our expectations. As a reminder, the $50,000 promotional upgrade price ended on October 1, and a price to upgrade is now $125,000. We do expect upgrade orders to continue but in a significantly slower pace. Through Early Access, HiSeq 2500s were shipped in Q3, and we are on track to commence full shipment of the product and upgrade in the next few weeks. This release will also include the ability of HiSeq to stream data to BaseSpace, our cloud computing solution. We continued to be pleased with the market penetration of MiSeq. MiSeq orders grew sequentially for the fourth consecutive quarter. Demand for this instrument is coming from across the globe, and we continue to see success with our trade-in programs. Despite launching our desktop sequencer platform 12 months later and at a higher price than the competition, we believe we have approximately 50% market share in the installed units and significantly higher share of the market revenue. More than half of the MiSeq orders continued to come from non-academic customers, the growing presence in the clinical and applied markets, as evidenced by the recently announced FDA selection of MiSeq for its foodborne pathogen identification program. This contract demonstrates the potential of the MiSeq to drive large multi-unit orders. MiSeq ASPs were down slightly in the quarter due to the continued success of our trade-in program. The average annualized pull-through per MiSeq was approximately $45,000. We remain focused on sales and marketing activities to drive continued growth in our share in this desktop market. Many MiSeq customers are experiencing up to 10G per run as a result of the upgrade we deployed this quarter. This no-charge upgrade is approximately 40% complete. Based on this performance, the breadth of enabled applications and MiSeq's ease of use we're very well positioned compared to the latest competitive entry in the desktop space, which is priced almost twice as high as MiSeq. Comparatively MiSeq, consistently generates more output and better data quality and has the easiest workflow. Combined with ease of storage, streamlined data analysis and data sharing features on BaseSpace, we believe we have the solution that best addresses the needs of the desktop market, including those customers with applied and clinical applications. Our FastTrack Services business had a very strong quarter. As Marc mentioned, we shipped more than twice as many samples in Q3 as we did in Q2 and achieved a new record for whole genome shipments. In addition, we continue to build our backlog, and ASPs for new orders significantly exceeds our expectations. In Q2, we announced the non-domestic release of our realtime PCR reagent portfolio, including a novel probe-based chemistry for gene expression analysis called NuPCR. We're pleased to report that we've now signed a major distribution agreement for our Eco instrument and NuPCR in the U.S., which we plan to announce in mid-November after we trained the distributor sales force and have the order processing systems fully in place. We've had some exciting development in our clinical business in the last several weeks, including the acquisition of BlueGnome. BlueGnome is a leader in the fields of cytogenetics and preimplantation genetic testing. It provides products for identifying genetic abnormalities associated with developmental delay, cancer and infertility. Its products include 24sure, a preimplantation genetic screening test for counting the chromosomes in a single human cell. This test is used by in vitro fertilization clinics and specialty testing labs around the world to increase the success rate of IVF while decreasing risks associated with multi-birth pregnancies. Clinical data have shown that PGS greatly improves IVF success, increasing pregnancy rates for women and reducing miscarriages. BlueGnome's portfolio also includes CytoChip, a powerful test for investigating genetic abnormalities associated with developmental delay or with complex leukemias. CytoChip is currently used by more than 200 labs across 40 countries as a first-line cytogenetic test, replacing traditional G-band karyotyping. BlueGnome is also well known for their solution-based approach to software design and assay workflows. Its BlueFuse software package is wrapped around their entire suite of cyto products and applications. The addition of BlueGnome as an Illumina company will enhance our presence in the cytogenetics market by allowing us to integrate our products into its software solution, while also providing a channel to IVF clinics. This acquisition is another building block toward our goal of leading the genomic based diagnostics market. In early September, we announced 2 other important steps along our clinical path. The first was the announcement of TruSight content sets, NGS panels that were developed in collaboration with recognized experts for specific genetic diseases or conditions such as autism, inherited cancer, cardiomyopathy, early childhood disease and rare disease. The second was an alliance with Partners HealthCare, integrating MiSeq functionality with their GeneInsight platform to streamline the analysis, interpretation and reporting of complex genetic test results. The combination of the content sets with new interpretation and reporting tools is a major step forward to further drive adoption in the clinic. We also remain on track to submit the MiSeq system to the FDA before year end. The individual genome sequencing service has continued to provide access to NGS, patients and their physicians the predominant focus on pediatric and cancer cases. We recently implemented the advances of the HiSeq 2500 into our CLIA services lab, enabling fast turnaround clinical whole-genome sequencing to be done in as little as 2 weeks. Our IGS service make it possible for healthcare practitioners to use data generated by whole-genome sequencing in the clinic, as the turnaround time is consistent with the demands of patient care. We believe that making our technology and platforms accessible, developing disease-specific content and creating robust interpretation and reporting solutions are key to the adoption of NGS in the clinic, and we continue to build on that understanding through deep collaboration with the communities driving this evolution. Yesterday, Illumina held an inaugural company-sponsored symposium called Understand Your Genome to explore best practices for deploying NGS in the clinical setting. This educational event brought together nearly 70 healthcare professionals, scientists, researchers, policymakers and consumers, including a Nobel laureate. During the event, these thought leaders explored laboratory and clinical considerations, implementing and deploying whole-genome sequencing in the mainstream clinical practice. Attendees who had their genome sequenced and annotated by Illumina's clinical geneticists were able to visualize their own genomic data on Illumina's MyGenome app for the iPad. We'll continue programs such as this to lay the groundwork for routine clinical use of whole-genome sequencing and to enable collaboration within the communities that will drive this transformation. In conclusion, we're very pleased with our execution this year and our outlook for the fourth quarter. We made tremendous progress with our market-leading instruments in both the high-throughput and desktop markets, and we're now field-upgrading both instruments to the latest specifications. We've announced a number of new products and collaborations and completed the acquisition of a remarkable business in the U.K. Our diagnostic story is starting to take shape, and we're extremely excited about the potential for NGS in the clinic. Thank you for your time, and we'll now open the lines for questions.

[Operator Instructions] The first question comes from the line of Tycho Peterson, JPMorgan. Tycho W. Peterson - JP Morgan Chase & Co, Research Division: First question just on the MiSeq utilization note [ph], I think it was about -- down about 10,000 quarter-to-quarter. Is that just the seasonality in the upgrade? Or was there something else there?

I think it certainly had a seasonality effect. I think the same effect, Tycho, that we saw on HiSeqs in the third quarter. It's not unusual for that to occur in Q3. I think we've also sort of bottomed out on the way our math works on the install base calculation that we do, and we're now really into the regime where we're trying to begin to push that number up again. I do think as well on MiSeq, we ought to be using sort of a rolling average calculation as opposed to single-quarter data points. Tycho W. Peterson - JP Morgan Chase & Co, Research Division: And then on BlueGnome, can you talk a little bit more about how you view the opportunity in cyto and how you see the existing synergies with your array business? And are you thinking more constructively about the outlook for arrays overall then in light of the deal?

Well, I think we certainly think that the cyto portion of the arrays is going to be a much stronger market opportunity for us. As you're probably aware, we're in roughly third place market share in cyto, about tied for where BlueGnome was. And I think, in combination, we now have the software solution that we didn't have with our standalone array technology. So we're very excited about implementing our arrays under the BlueFuse software capability. They also have a sales force that's fully directed towards these IVF laboratories and cytogenetics labs, which was an area that our sales force was not spending enough time on. And we are going to have a dedicated sales force inside the BlueGnome entity that will be focusing on this particular market. So we're very optimistic about our ability to significantly increase our share in the cyto market going forward and continue now to pour more technology into advancing the state of the art of our cyto products. And as we've mentioned previously, we plan to submit the iScan system and our cytogenetic arrays to the FDA shortly. Tycho W. Peterson - JP Morgan Chase & Co, Research Division: Okay. And then just one last one, are you able to comment on book-to-bill in backlog? How are you thinking about managing MiSeq backlog in the back half of the year?

We don't routinely comment on that. I know we did for 2 quarters in a row earlier in the year, but we're not going to routinely give out the book-to-bill ratio.

Our next question comes from the line of Doug Schenkel of Cowen Group.

Was the sequential decline in HiSeq pull-through a bit more material than you expected? And what are you baking into full year guidance? Are you expecting a sequential improvement in HiSeq pull-through in Q4?

I guess, we didn't find it particularly alarming as we've been looking at that math over the last couple of months. One thing was pretty apparent, and that is that we have significant catch-up in reagent shipments in the second quarter. We had a pretty substantial backlog, particularly in the RNA area of kits that we really got focused on in Christian's organization. And they did a great job of sort of getting that aged backlog shipped out. And so, I think there was a little bit of peaking -- peak's not the right word, but a sort of a catch-up in Q2 that wasn't part of the real run rate that underlies the HiSeq utilization. So when you look at the curve that way, the number didn't look particularly unusual, particularly in combination with the summer months, particularly in Europe and Asia where there's lots of vacations going on. So there's nothing startling about that number for us. As we look forward, I think we're going to continue to work hard to push the number up. There's reasons because of the 2500 rapid run mode that might be a positive influence on the number. The pricing is going to be pretty much fully baked in, in Q4 and totally baked in, in Q1 of next year. So that will be a positive up draft on the number as well. We don't give specific guidance around that number. So we're not going to comment on exactly where we think it might go. We'll sit and watch it as you guys will.

Okay. And one other question. Factoring in the BlueGnome contribution, I believe you said $5 million in Q4. You effectively reduced your full year expectations by that same amount. Recognizing this is a fairly trivial 0.5-basis-point decrease in full year growth, what's changed? What's coming in a little light of expectations? It sounds like it's not that pull-through in HiSeq. So is it related to -- is there something specific that you'd like to highlight? Is it just the environment? Is it the timing of the NuPCR distribution trading, timing of the HiSeq 2500 launch? Anything that we should be thinking about that's new in terms of how you're thinking about the outlook.

The first comment I'd make, Doug, is that this new guidance range is really a narrowing of the range we gave before. So we haven't truncated our guidance in any way from what we've given. I think what you're comparing to is the consensus estimates that are out in The Street. And you're accurate that if you factor BlueGnome in, this is a slight reduction from what the consensus numbers look like. The way to think about this really is that we're going into Q4 cautiously, because we have an election coming up. We have some potential disruption in November and December. Based on what happens with the election, the looming fiscal cliff is just difficult for us to predict what customer behavior may be in light of those changes. We hope it remains normal as we saw in Q3. And as a backdrop, I think we would comment that we've seen no erratic customer behavior based on macro conditions whatsoever in the third quarter, unlike what we saw in Q3 of last year. So our guidance really reflects some caution going into this. The range also reflects both extremes of the R&D tax credit, either getting approved or not getting approved, is included in the range of those numbers, as well as $5 million from BlueGnome in the numbers.

Yes, I think that's about right.

Our next question comes from the line of Derik De Bruin of Bank of America.

Can you just talk about what's -- have you seen any change in the competitive dynamics in the market now that there's a new NGS box that's sort of out there? Has -- have your -- what's going on in terms of how your competitors approaching customers? Is -- just some color on the market, basically?

Well, I'd say certainly in the high end of the markets. Let's take that first. Things are going really, really well with 2500 and the 2000 sales as we indicated in the script. We had very strong performance that exceeded our expectations in the high end of the market. And we essentially have no competition in that part of the market. In the desktop, I'd say, if anything, our competitive position has continued to improve there. The new launch of a competitive instrument into the field is from any customer we've talked to not performing to specifications. And so we're not seeing our customers get distracted by that system. We're seeing already some pretty heavy discount offers of that system into the marketplace. And our MiSeq product continues to perform really, really well. And we're upgrading the output in the system, and the customers are continuing to get great data with BaseSpace enabled for storage and for data sharing. So all of the attributes, ease of use of our system continue to make it very, very competitive. And certainly, the output ranges that customers are achieving with this exceed the output ranges of any competitive systems in the desktop market.

I think the other thing I'd add, Jay, is -- going back to MiSeq, the out-of-box quality has improved dramatically, and that gives people a lot more confidence to buy their second and third system. I think we're seeing some orders like that. And then on the HiSeq, the good news is the instrument sales were really strong, which bodes well for a long-term consumable pull-through and really shows our leadership there at the high end.

And so -- and what happens now that it looks like BGI is going to be purchasing another competitor of yours. I mean, what if that happen -- what did -- how does that change your relationship with BGI? I mean, how much of their work is being done as whole-human genome sequencing?

We don't know exactly. It's probably in the range of 10% to 20% would be our guess. As I said, we don't know that number for sure. We're in discussions with BGI to get a sense of where the relationship may go. As you're aware, that acquisition is not yet completed, and so they don't own the entity as yet. And it will be at least several more months before that happens, if it does. And so, at that point, we'll engage very directly with BGI in a conversation about how the relationship evolves.

Our next question comes from Amanda Murphy with William Blair. Amanda Murphy - William Blair & Company L.L.C., Research Division: So I just had a follow-up to Doug's question first on the HiSeq consumables. So you talked really to 3 dynamics in the past. And you mentioned obviously pricing, but I'm curious if you can provide a little more color on underlying utilization. Obviously, you're working to do some things to improve the front and the back end. Are you seeing any improvement at all in underlying utilization? And then also sample prep, I know you've mentioned some market penetration rates there. Is that -- slowly picking up? Or can you can talk [indiscernible] those 2 dynamics, that'll be helpful?

Well, I guess there's not a lot more to say, Amanda, about it. I mean, clearly, at the highest level, the actual measure we used went down a little bit sequentially. We think that the factors that we mentioned were the key contributors there that there was lower utilization, particularly in July and August. As we all are aware, Europe is pretty slow during those months as are some of the countries in Asia. And so it's not a surprise to see overall consumption down a little bit and it's usually that seasonal effect that we see every year. And what else can we say about it? I think we're continuing to work hard to push it up. Fundamentally, customers are continuing to buy these units. And so that is a positive indicator that they're fully utilizing the ones they have. The largest proportion of these are going to existing customers as opposed to new customers. And so that, again, reinforces the fact that the customers are filling up the ones they've got and moving on to another one.

And on the sample prep side, the sample prep business has overall been nicely growing. Particularly, if you look at year-over-year comps over the whole year, we've had a nice uptake in that, and it's really been -- a key focus of ours is to get a full suite of products out there which we've been able to do. And so we've seen a nice addition to business there. And I definitely think we've been taking market share in that area over the last couple of quarters. Amanda Murphy - William Blair & Company L.L.C., Research Division: Okay. And I just had a couple of questions on the clinical market. So if you look at the translational clinical market, obviously, it's an early market. I'm just curious if you can talk to -- I guess, what are people using the MiSeq for in the clinical space? Who is really buying it? And could you -- do you expect the order flow to continue to tick up over time? Or could we see a pause just as you worked through some of the early adopters in that space?

Well, I guess, I'd answer the last part first and say that we do expect that business to continue to accelerate. The largest customers that we currently have in that marketplace are a set of companies that are doing T21 testing. So we have one here in San Diego and 4 others in other parts of the U.S. and one in Europe as well. And those companies are ramping up to pretty high test volumes. So this is a clinical set of customers that are beyond the assay development phase and are ramping up to quite high volumes, and they're becoming a significant customer base for us. The other customers we're focused on are going to be adopters largely of targeted panels of one sort or another. So many of them are doing panel development at this stage. More and more of them, we think, will begin to use the TruSight panels that we've released. Because it gives them a great baseline to start with, and we may take inherited disease panels as an example and then add a few genes that they think are particular -- maybe they have intellectual property around those genes or maybe they think they're somehow relevant to the populations that they tend to work with. And they would be adopters of those panels. And so I think that's the kind of application that you would see. The type of users run the gamut from the large clinical laboratories, the LabCorps, Quests and Mayos of the world; the cancer centers; down to small regional laboratories to university hospital labs. So it's a very wide range of customers. One trend we've started to see this quarter, which I think is an interesting one to begin to think about, is the number of repeat orders of MiSeqs. Because what we have largely been doing is seeding the market with most customers buying a single unit. And we saw this quarter the incoming order rate, 20% of the incoming orders were on multiple unit orders. And so that's a change from where we have been over the past couple of quarters, so a very positive trend there. Amanda Murphy - William Blair & Company L.L.C., Research Division: And do you have any visibility into -- of end-user demand, if you think about physicians and how they're using these target panels and payors in terms of how they're paying for them? Again, it's early, I know, but I'm curious if you have any insight there into the end-user demand dynamics.

I think it's different in every lab and every market and every test. So it's a little hard to generalize that. What I would say though is that there are existing CPT codes -- these are going to change next year. But these [ph] CPT codes are applicable to most of these tests. If you were doing particular genetic tests on our platform and that test has already been done using some other methodology, very often the lab can apply for a reimbursement based on that CPT code, at least for portions of what they're doing: sample prep steps, PCR amplification, data analysis. And so many of the existing codes are applicable.

Our next question comes from the line of Ross Muken, Deutsche Bank.

I just wanted to go back to sort of the consumable pull-through on the HiSeq. I mean -- so I wanted to talk about the consumable pull-through again. I just wanted to sort of understand the trajectory. I mean, the implication, what you were saying, was that we sort of had a weak July, August. And then it sort of -- it improved into September, just based on the sort of summer holiday commentary. I mean, can you give us directionally, if you can, any of the color on what utilizations look like in September, October and kind of the implication vis-à-vis an acceleration for 4Q, in terms of what's implied in guidance there?

Well, I -- we don't actually do the calculation on a monthly basis, Ross. And so -- and it would be a meaningless number, to be honest, to do it on a single-month basis. So I tend to make you think about a longer time frame. Don't look at it on a narrower time frame, such as one single month. I think if you normalize what happened over the last 4 quarters, and by that, I mean you pull Q2 down some because of the catch-up on the backlog, and you pull Q3 up a little bit by -- if you were to add that -- have smoothed that backlog out. There is an increasing slope to that line. And as I said, we're not going to guide specifically to what we think in Q4. But there's a handful of factors that would cause us to believe that it could push up. Again, think about this over a multi-quarter time frame, not one quarter and one data point being a particular trend line. But over the next couple of quarters, in general, we would expect a positive slope to the line.

How are you thinking about pricing in that market on HiSeq consumables based on sort of your own plans in terms of the upgrades, based on the competitive products, whether it's the Proton or anything else? How do you think about sort of the transition there of that base in terms of how other consumable pull-through is going to trend, perhaps more so on an 18-month basis?

Well, we had a pretty significant price increase this year on the reagents. Some of our customers avoided that by putting standing orders in and by doing their web ordering. That will fully expire by the end of this year. And we'll be looking at pricing for next year. I think, certainly, we wouldn't expect that we'd be raising pricing to the extent we did last year. There might be a more nominal nudge in pricing going into 2013. As I said earlier, we really are the company that can sequence human genomes at this scale. And so I don't think any competitive systems are going to have a particular impact on our pricing. And in general, the way we tend to do this is to sort of nudge pricing and then whenever we make a performance increase, we would make a bigger step change in the pricing, similar to what we did with the 2500 kit, as an example.

And just lastly, housekeeping. On the other expense line, what was the delta there? I mean, that came down a couple of million sequentially. What sort of the underlying in that?

Yes. Ross, it's Marc here. I mean, there's a couple of factors that go into that. FX has a minor impact. In addition, we realized some gains on the sale of some investments to fund the BlueGnome acquisition.

Our next question comes the line of Jon Groberg of Macquarie Capital.

Just 2 ones for me. One is following up on the pricing, Jay. Can you maybe -- would you maybe want to say this year what percent of your revenue growth you think you expect to come from price increases, given you said some of your customers were able to avoid it?

I don't have that number in front of me, Jon. We could probably do that math and give back to you with some rough idea of what that is. But we have to do some realtime math here that we probably shouldn't do.

Okay. And then another quick one. I just want to make sure I was doing the math or maybe you guys can help me out because something is probably wrong. I think Marc said $82 million in instruments, which is about $10 million year-over-year, and that most of that was driven because of the MiSeq. But if you do -- you assumed some discount on the MiSeq, there's still not that many systems. Can you maybe just talk about what happened to MiSeq? Was it up -- the revenue is up sequentially, down sequentially? Just trying to understand that one.

Yes. Jon, it's Marc here. The -- so the year-over-year -- don't forget we only started shipping MiSeq in the third quarter of last year. So it was only a partial quarter of shipments. So that's really why we made the comment that MiSeq is driving the year-over-year increase in instruments.

It's actually only a fraction of 1 month, to be honest, in the third quarter of last year.

Right. Yes, right.

But sequentially, were MiSeq instrument revenues up or down?

We haven't disclosed, or we're not disclosing the breakdown between MiSeq and HiSeq sequentially. But as Jay said, we've seen 4 consecutive quarters of order growth in MiSeq.

Our next question comes from the line of Bill Quirk, Piper Jaffray.

First off, Jay, I was just hoping to get your thoughts on ENCODE and the potential effects, both kind of negative on the array business and positive from a sequencing standpoint and how you look at that. And then also -- and I certainly realize this is just recently published. But have you seen any effects, thus far, in behavior?

I guess, first thing I'd say is that, I think ENCODE has the potential to certainly help our sequencing business over a period of 1 year or 2 years. And the reasons I think are pretty clear that there is tremendously valuable information outside the exons, and the science needs to get at that information and study it at large scale. The best way to do that, of course, is to sequence the entire human genomes. And so that's going to be a key driver of the sequencing business. I do think, however, there are some opportunities in arrays as well. Because once we understand the specific points in the genome that matter outside the exons, whether these are switches or something going on in the regulatory regions, we have the ability to go back and put those on arrays. And so there could, in fact, be some follow-on content arrays that could be quite interesting as a direct follow-on to what was learned in ENCODE and what we'll learn over the next year or so. So we're pretty optimistic about that really beginning to drive the business and also perhaps to increase the funding compared to what it would've been otherwise. Because I think there's significantly renewed vigor around the whole idea of where in the genome we may find some indicators that account for this sort of dark matter or lack of explainability of the inherited diseases. And so I think the funding agencies may have renewed optimism about increasing the funding levels to go after that.

Got it. And then just 2 quick ones for me. First off, service continues to pick up nicely and certainly understand that we're seeing more orders for whole genomes. But we're also a little over 1 year away from you guys shipping a lot of instruments vis-à-vis HiSeq and then obviously, coming up on the MiSeq launch. And so how should we think about that in terms of new service contracts coming on and driving that?

Yes, that -- I mean our service contract business has continued to grow very nicely. And so I think you should continue to expect that number to rise. And our margins are reasonable on that business. So it's a strong contributor for us.

Okay. And then last one for me, Jay, is just regarding the TruSight panels. Obviously, it looks fairly interesting, particularly from a clinical standpoint. Can you talk to us about when we might see some data to -- or some more clinical data rather to support those and basically see those becoming, as perhaps, bigger diagnostic targets?

I think the real value of the data coming off of those panels will be delivered by our customers more than by us. And so I think you might see some data at AGBT perhaps. We'll start shipping the first of these in Q4. And what really matters is when the customers apply these panels to large clinical cohorts. We're not doing any clinical trial with these panels. These are purely content panels of known content. And so there's no reason for us to do pure clinical work with these. We've done all the analytical work to ensure that they get great performance and deliver great performance. But what, at the end of the day, is going to matter is what customers can do clinically with them.

Yes. And Bill, I want to remind everyone that these are RUO panels only.

Our next question comes from the line Isaac Ro, Goldman Sachs.

I just wanted to see if we can get an update on the BaseSpace App Store. And the reason I ask is I know a lot of -- a large percentage of MiSeq users have been connecting to the service and wondering if you could share some feedback on how important you think that is in helping you win placements and maybe how that would change once the App Store is live.

Yes. What's made me really optimistic about this. The adoption rate of BaseSpace is high, and I think there's a lot of anticipation around launch of BaseSpace here in Q4. And we're going to have some demonstrations at AGBT -- I'm sorry, ASHG, coming up next week or week after?

Week after next.

Week after next, yes. And so we will have some specific applications up and running in the App Store during that show. And so we'll have some demonstrations of it, and then it's going to take us 1 quarter or 2 to begin to get a sense of how often they're used. But we're really pretty optimistic about this. There's been a significant number of upload runs. So...

Yes. We're over 15,000 uploads for -- people uploading their data on the App Store. We're going to launch with nearly a dozen different Apps. So it's pretty exciting.

15,000 uploads.

Okay. And then if I can just maybe, on the informatics topic, kind of switch gears to the sort of utilization outlook, given the consumable pull-through this quarter. Can you maybe comment on whether or not the slowdown in pull-through was in any way tied to informatics bottlenecks from any of your customers? Do you think there's any chance some of these people are slowing down their usage a little bit, so they can kind of process the data?

We don't see any signs of that. In fact, I think the work we've been doing on the software side in BaseSpace and other places probably means that the processing of data today is easier than it was 1 year ago.

Our next question comes from the line of Dan Arias, UBS.

Just to the points on the multiple system orders. I'm wondering if you can give a sense for what percentage of your MiSeq user base actually has more than one instrument at this point.

I don't know that number exactly. But it's probably in the range of 20% to 25% have more than one unit.

Okay. And then just on the outlook for arrays, maybe excluding the cyto products, would you be willing to parse out the growth trajectory for custom content and focused content versus whole-genome products?

Yes. I think we're happy to do that in general. We're not going to quote exact growth rate. But I think you could expect the custom and focused products to continue to have some growth. The exome product has done really well for us and the new HumanCore products. We have some reasonably high expectations for as well. That will be offset by what we think will be a continued reduction in the GWAS part of that business. So we're not optimistic about the continuing drive to whole-genome studies. And so I think those 2 will largely offset. And the way we think about the array business in total is sort of flattish to maybe up a little bit.

And just thinking about M&A lastly, Jay, is there content out there on the diagnostics side for sequencing, specifically, that you think is important enough to acquire versus maybe partner when you go to put the pieces of that program together?

There certainly is some. There's not a lot. And what we're finding and I think this is becoming realized, for example, by the pharma companies, is that any specific marker in a gene or 1 or 2 markers may in fact have some connection to how somebody responds to a drug or some particular disease effect of that gene. But there could be many other markers in that gene that have the same deleterious effect. And so the problem we're finding in licensing content is that everybody believes there are 1 or 2 markers that are worth $5 million or $10 million to create the world's most enormous diagnostics business. When in fact, it's maybe going to be 100 different variants in that gene that could have some influence on how somebody reacts to a drug or what disease they might get. And so I think that's an inherent structural problem, we think, in going out to acquire any meaningful content. So there's obviously some exceptions to that of very specific markers. But in general, that's what we find. It's part of -- I should say it's part of the reason that we see pharma companies going more and more to sequencing the entire genes as companion diagnostics as opposed to doing single-point mutation analysis.

Our next question comes from the line of Amit Bhalla, Citi.

I wanted to just get a little bit more clarification on your comment, Jay, about the MiSeq promotion than ASPs. Can you talk a little bit about that the magnitude of down slightly means? And how you're thinking about promotional pricing going forward for MiSeq?

It really -- Amit, it's really what Jay was talking about there -- it's Christian, by the way, is that we were -- what we have is we have a trade-in program where we can trade in whole-genome analyzers or competitors' platforms, and they can get -- customers can get a pretty significant discount off of the MiSeq. And so, in any given quarter, the number of those trade-ins impacts the ASP. And so that's kind of what's -- one of the primary drivers of why the ASP is different from quarter-to-quarter.

And we really characterize that as a trade-in program as opposed to a promotion.

Okay, so you're not saying that there's any other pricing discounts outside of the promotions or the trade-in program that you're talking about.

No. Well, there's some discount. I mean, there's…

Well, there's always multi-unit discounting. But our discounting on MiSeq is quite nominal, to be honest. We've held quite close to the list price. But there is some impact from the trade-ins.

Yes. And I wouldn't characterize -- we used the word slight, because it wasn't very significant in the quarter.

Okay. And then the second question, I think you mentioned in the prepared comments that some of the users with the new chemistry are getting 10G of output with MiSeq. Can you talk about room for continued improvement on MiSeq output, whether it's through bigger flow cells, clusters? Can you talk just a little bit about room for improvement beyond the 10G that some people are seeing?

You should think about the MiSeq much in the way that we thought about HiSeq historically in GAs before the HiSeq that these are platforms that continue to have significant headroom. We continue to push the technology across many, many different dimensions, cluster densities, cycle time, accuracy, improved gentomology [ph], faster image extraction, full cell sizes. All of those factors have historically played into our ability to continue to improve throughput and decrease the overall run times in the instruments. And as we've said, I think, on many occasions, we're a long way from running out of headroom across that set of parameters.

Okay. And just a quick clarification for Marc. I think I may have missed it, the increase in DSOs and accounts receivable. Can you go into a little detail there?

Amit, it's really around the timing of shipments and when they occur in the quarter. If you kind of do a look back at when we build and kind of look at the DSO on that basis, it's really a very immaterial change quarter-over-quarter.

Our next question comes from the line of Daniel Brennan, Morgan Stanley.

Just on the clinical opportunity. You made certainly several moves with BlueGnome and Partners. But I'm just wondering, as you look out into 2013, what kind of additional staffing or investments or partnerships? How should we think about your further needs to optimize your opportunity there?

Well, certainly, there will be continued investments in the diagnostics part of our business and the translational parts of our business. This will be for some clinical trial work in the diagnostics area. We are going to begin to put some test through the FDA beginning in the next few months, and that will continue through 2013. And each of those has probably in the range of $1 million to $3 million or $4 million worth of clinical trial expense associated with them. We'll continue to invest in building panels, both for RUO and for true diagnostic-type applications. We will continue to build the sales force there, particularly in the BlueGnome side of the business. Because we think we have a fantastic technology now with a fully integrated solution. And so we need to ramp a little bit on the commercial side of that. So I think you will see continued investments in '13. But you'll see revenue to match as well.

Okay, great. And then maybe just switching gears just to MiSeq. Just kind of wondering how you think about kind of where you are with the penetration of your addressable opportunity there, both on the research side and the diagnostics side of the market.

I think, in both cases, it's very early. Many -- most of our customers still have only one unit, and most of those customers have the potential to take several, if not significant numbers of these units as they deploy assays in the production-type environments. And that's particularly true in the translational and clinical-type settings. Should we get an FDA-approved box, which is clearly one of our goals, I think that gives us a very strong initial competitive advantage and a reason that a whole new set of labs would begin to transition to using sequencing as part of their routine usage in a clinical environment, because now they have an FDA-approved box and/or an FDA-approved test. And so I -- so we begin to really push into a whole another new set of customers. There's many applied markets that are opportunities for us going forward, if you look at what we did with the FDA for foodborne pathogen testing, and that's really a brand-new market for us all together. And in places like that, in forensics, there's huge untapped opportunities for us.

And ladies and gentlemen, that concludes the Q&A session. I'd now like to turn the call back over to Rebecca Chambers for closing remarks.

Thank you. As a reminder, a replay of this call will be available as a webcast in the Investors section of our website, as well as through the dial-in instruction contained in earning -- today's earnings release. Thank you for joining us today. This concludes our call, and we look forward to our next update following the close of the fourth quarter and fiscal 2012.

Ladies and gentlemen, that concludes the presentation. Thank you for your participation. You may now disconnect. Have a great day.