Illumina, Inc. (ILMN) Q4 2010 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the Fourth Quarter 2010 Illumina Inc. Earnings Conference Call. My name is Yvette, and I will be your operator for today. [Operator Instructions] I would now like to turn the call over to Mr. Peter Fromen, Senior Director of Investor Relations. Please proceed, sir.

Thank you, operator. Good afternoon, everyone, and welcome to our Fourth Quarter 2010 Earnings Call. During the call, we will review our financial results released today after the close of the market, offer commentary on our commercial activity, and provide financial guidance for 2011, 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. Presenting for Illumina today will be Jay Flatley, President and Chief Executive Officer; and Christian Henry, who is our Senior Vice President and General Manager of Life Sciences, as well as our Chief Financial Officer. This call is being recorded, and the audio portion will be archived in the Investor 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 be protected under the Private Securities Litigation Reform Act of 1995. The 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 the results to differ, we refer you to the documents that Illumina files with the Securities and Exchange Commission including forms 10-Q and 10-K. Before I turn the call over to Christian, I want to let you know that we will participate in the Cowen Healthcare Conference in Boston the week of March 7, and in Barclays Global Healthcare Conference the week of March 14 in Miami. For those of you unable to attend any of the upcoming conferences, we encourage you to listen to the webcast presentation, which will be available through the Investor Relations section of our website. With that, I'll now turn the call over to Christian.

Good afternoon, everyone, and thank you for joining us today. During today's call, I will review our fourth quarter financial results and discuss our financial guidance for 2011. Jay will then provide an update on our commercial progress and the state of our business and markets. In the fourth quarter, we recorded $261 million in total revenue. This represents growth of 45% over the fourth quarter of last year. The product revenue was $246 million, representing 47% growth over the same period in 2009. Our revenue growth was driven primarily by our sequencing products but also by solid year-over-year growth in our Microarray business. We generated total consumable revenue of $132 million in Q4, which was up 26% on a year-over-year basis. This growth was largely the result of our expanded install base of sequencers, which generated annualized consumable revenue per system at the high end of our range of $150 million to $200,000 per system. On the HiSeq system, consumable shipments in Q4 were over 350,000 per system on an annualized basis. The significant pull through of the HiSeq was largely offset by decreases on the Genome Analyzer, as customers transitioned to the new platform. As HiSeq becomes a greater percentage of our installed base, we expect the aggregate consumable pull through to increase accordingly. On a sequential basis, consumable revenue was roughly flat as growth in sequencing was offset by a decrease in array consumables. Annualized consumable pull through across our installed base of microarray scanners was at the high end of our targeted range of 400,000 to 500,000 per system, but due to the timing of shipments, revenue was down sequentially. However, orders for microarray consumables were at record levels in the fourth quarter. Instrument revenue for the quarter was $110 million, up 81% over the fourth quarter of last year and 25% over last quarter. In both cases, the growth in instrument revenue was largely due to the success of the HiSeq 2000. Increased production of the HiSeq in Q4 allowed us to ship a record number of systems. With our scale up now complete, we will significantly reduce delivery times over the first half 2011. An additional outcome of this scale up was our ability to produce and ship 29 HiScanSQ systems in Q4. HiScan shares many of the same components with HiSeq, and the combined demand greater challenges in delivering on the HiScanSQ orders. Breaking this bottleneck resulted in revenue growth for micro instruments of over 100% year-over-year and over 80%, sequentially. Services and other revenue, which include genotyping and sequencing services as well as instrument maintenance contracts was $16 million, compared to $13 million in Q4 of last year. The primary driver of year-over-year growth was the increase in maintenance contracts for our growing installed base of sequencing systems. Before discussing our gross margins and operating expenses for the quarter, I'd like to note that we recorded a pretax amount of $20 million associated with non-cash stock-based compensation. This impacted our EPS by a tax adjusted amount of $0.09 per pro forma diluted share for the quarter. As a reminder, we now include this expense in our presentation of the pro forma net income and earnings per share. However, in our discussion of gross margin, operating expenses and operating margin, I will highlight both our GAAP expenses, which includes stock comp expense, and other non-cash charges, and the corresponding non-GAAP figures. I encourage you to review the GAAP reconciliation of our non-GAAP measures included in today's earnings release. Total cost of revenue for the quarter was $95 million compared to $55 million in the fourth quarter of 2009. The Q4 2010 cost includes stock-based compensation expense of $1.6 million compared to $1.3 million in the prior-year period. Excluding this expense and $2.3 million associated with the amortization of intangibles, non-GAAP gross margin was 65.1%. This compares to 67.8% last quarter and 71.2% in the fourth quarter of 2009. There are two key factors that contributed to the sequential decline in gross margin in the quarter. First, the mix of revenues was much more weighted towards instruments compared to the prior quarter. As you may realized, instrument gross margins are generally lower than consumable gross margins. Second, the gross margin of instrument revenue was, as anticipated, significantly lower than the prior quarter due to a greater number of HiSeq shipments associated with the Genome Analyzer trade-in program. Shipments of HiSeq systems under these trade-in programs were approximately 75% complete at the end of Q4 2010. And as such, we expect the dilutive impact of these trade-ins to decrease substantially in the first quarter of 2011 and be a non-factor beyond the first quarter. R&D expenses for Q4 were $46 million compared to $40 million in the comparable period of 2009. These numbers include $7 million and $5.6 million respectively in non-cash stock compensation expense. Excluding this expense and $0.9 million of accrued contingent compensation in both periods, R&D expenses were $38 million or 14.4% of revenue compared to the prior year R&D expenses of $34 million or 18.8% of revenue. SG&A expenses were $62 million compared to $49 million in the fourth quarter of 2009, including stock compensation expense of $11.3 million and $9.6 million, respectively. Excluding stock comp expense, SG&A was $51 million or 19.4% of revenue compared to $40 million or 22.1% of revenue in the prior-year period. GAAP operating profit for the quarter was $69 million. Excluding the expenses outlined earlier and a gain recorded for the change in fair value of contingent consideration related to a prior acquisition of $10 million, our non-GAAP operating profit for the quarter was $81 million or 31.3% of revenue compared to $55 million or 30.3% of revenue in the fourth quarter last year. GAAP interest and other expense in the fourth quarter included a $13.2 million impairment charge related to a cost method investment. This charge is related to an investment in an entity unrelated to our sequencing business. Although we continue to believe in the value of the underlying investment and our associated research collaboration, uncertainty regarding future funding of this entity required us to impair this investment as of the end of the fourth quarter. Additionally, the GAAP figures include a $5.4 million non-cash interest expense charge associated with our outstanding convertible debt. Excluding the impairment and non-cash interest expense, pro forma interest and other income was $0.7 million. Foreign currency exchange rates did not have a material impact on our business during the quarter. Our non-GAAP tax rate for the quarter was 34.3% compared to 35.5% in the last quarter. The decrease was primarily due to a benefit from the retroactive extension of the U.S. R&D tax credits recorded in Q4, partially offset by cost associated with IP restructuring that had a negative impact on the Q4 tax rate. We believe this restructuring will have a positive impact on our future tax rates. We reported GAAP net income of $38 million or $0.25 per diluted share compared to the net income of $12 million or $0.09 per diluted share in the prior-year period. Excluding the items identified in our press release and net of pro forma tax expense, non-GAAP net income was $41 million or $0.29 per pro forma diluted share compared to $27 million or $0.21 per pro forma diluted share in the fourth quarter of 2009. I'd like to note that our GAAP and our non-GAAP net income per share were negatively impacted by approximately $0.02 due to increased stock-based compensation and share count from the Q4 rise in our stock price. During the fourth quarter, we generated $81 million in cash flow from operations. We used approximately $12 million for capital expenditures. This resulted in free cash flow of $69 million. This compares to $55 million of free cash flow in the fourth quarter of last year. Free cash flow benefited from another strong quarter of collections, which yielded a DSO of 58 days as compared to 81 days in Q4 of last year and 65 days last quarter. In addition, we used $28 million to repurchase our common stock. As a result, we ended the quarter with approximately 894 million in cash and short-term investments. Inventory increased approximately $12 million in the quarter primarily due to the scale of the HiSeq 2000 instrument manufacturing line. Depreciation and amortization expenses for the quarter were also approximately $12 million. Before we discuss specific annual financial guidance for 2011, I'd like to expand on the financial impact of the lease of our new corporate headquarters. Due to the downturn in the real estate market, we were presented with the unique opportunity to relocate our corporate headquarters to a campus in San Diego with significant expansion flexibility. In December, we completed a transaction that enabled us to meet our long-term expansion needs, and is expected to be an approximately cash flow neutral with respect to our existing and planned lease obligation at our current location. We expect the transition to the new facility beginning in the fourth quarter. Coupled with this transaction, we expect to record a onetime non-cash charge of approximately $30 million related to our remaining lease obligations. In addition, we expect to incur additional expenses associated with accelerated depreciation and a brief period of double rent. We intend to exclude these items from our pro forma results. All of this was accounted for in the preliminary guidance that we provided in January. I will now provide our financial guidance for 2011. Our guidance includes the integration of the acquisition of Epicentre Biotechnologies, which closed last month and excludes certain charges associated with the move of the company's corporate headquarters, incremental interest expense associated with the company's convertible debt, the amortization of intangibles and acquisition-related charges. As a reminder, we now include non-cash stock-based compensation expense in our earnings per share guidance in addition to our discussion of net income and the corresponding per share amounts. For additional details, please refer to the table in our earnings release that reconciles our non-GAAP guidance to the related GAAP figures. We expect revenue growth for the full year of 2011 of approximately 20% from 2010 revenues of $903 million. We expect gross margin to improve over the course of the year to 70% for the full year. We anticipate non-GAAP earnings per share to grow more than 30% from 2010 non-GAAP earnings per share of $1.06, and we expect to incur stock compensation expense of approximately $101 million or a tax adjusted amount of $0.49 per pro forma share. We expect full-year weighted average diluted shares outstanding for the measurement of pro forma amounts to be approximately 145 million shares. At this point, I'd like to turn the call over to Jay for some remarks on our commercial activity during the quarter before we begin our Q&A session. Jay?

Good afternoon, everyone, and thank you for joining us today. I'm pleased with our operational and financial results for the fourth quarter. We delivered strong revenue growth as we continue to scale the production of our sequencing instruments and consumables. This enabled us to reduce lead times for the HiSeq instruments, and cleared a backlog of multi-visit launch promotions, paving the way for gross margin improvement in 2011. The market for the HiSeq system continues to be robust. And as a result, Q4 orders were at record levels, excluding the BGI [Beijing Genomics Institute] order recorded in Q4 2009. Total revenue grew 45% year-over-year, and we exited the year with the largest backlog in the company's history. In the field, the HiSeq transition continues to go extremely well. Customers are consistently generating runs yielding throughput well above our product specification. To meet demand, we increased HiSeq manufacturing capacity by approximately 30% over Q3, and now believe we're in a position to materially reduce customer lead times. As Christian mentioned earlier, we're seeing encouraging uptake in HiSeq consumables as more customers place instruments in production and increased utilization. In fact, in the fourth quarter, the incoming order rate for HiSeq consumables was equivalent to that of the Genome Analyzer. Total Q4 microarray revenue grew sequentially and year-over-year, lead by strong results in array instrumentation, partially offset by a sequential decline in array consumable shipments. Total consumable orders and shipments were particularly strong in the third quarter given the end of the NIH fiscal year, and consumable orders continue to grow in Q4 adding to the array backlog. GWAS shipments in Q3 were strong as a result of the first full quarter of Omni 2.5 availability but down sequentially, in Q4, as these customers focused on execution of their rare variant pilot projects. We expect to see results from these studies around midyear, and believe that some customers are waiting for results before commencing studies on our rare variant arrays such as the Omni 2.5 and the upcoming Omni 5. Looking at 2011, we anticipate that positive results from the rare variant pilot studies will renew demand for GWAS. Overall, orders for total array consumables grew 7% sequentially, and 14% year-over-year. We are in development of the Omni 5, which is expected to increase genomic coverage and incorporate additional rare variant content from the 1,000 genomes project. We plan to ship the Omni 5 this summer. In early January, we enhanced our iSelect custom array product line introducing three new formats that allow complexity up to 1 million custom markers. In addition to increased complexity, customers have the option to augment their content at any time after they design the initial chip. We feel this is a highly compelling offering for researchers who want to add new content to their association studies produced by ongoing sequencing projects. We generated record revenue in microarray instrumentation in Q4, as we were able to reduce our backlog of HiScan and HiScanSQ systems. In fact, array instrument revenue more than doubled from Q4 last year and nearly doubled from last quarter. Orders for HiScan and HiScanSQ also grew sequentially, as customers showed continued demand for the platform's flexibility to perform sequencing and array applications on a single instrument. As most of you know, last summer, we announced the acquisition of Helixis and at the same time, launched the Eco Real-Time PCR System. In Q4, Eco orders tripled, and shipments nearly tripled over the third quarter. We've received positive customer feedback on the system's high-performance and attractive price points, and continue to focus on scaling Eco manufacturing with the expectation of marching market demand by the middle of 2011. We're in the process of moving Eco manufacturing to Singapore and expect to place it under GNP by the end of 2011 in anticipation of increasing diagnostic applications. Turning now to our sequencing business. We had another outstanding quarter. Total sequencing revenue grew by over 70% from the fourth quarter of last year and continues to be supported by broad adoption of the HiSeq 2000. Excluding genome centers, over 75% of the HiSeq shipments included Genome Analyzer trade-ins. This mix of shipments was the primary contributor to lower gross margins compared to Q3. Looking to Q2 2011, we expect the impact of the Genome Analyzer trade ins on gross margin to taper off beginning in the first quarter. The breath of the market opportunity for HiSeq has continued to exceed our expectations. Notably this week, we passed through 1,400 peer review publications citing SBS chemistry. Nearly 90% of Q4 HiSeq orders came from outside of major Genome centers. By simplifying the sequencing workflow, increasing accuracy and dramatically lowering the cost per Genome, HiSeq is truly redefining the trajectory of sequencing and expanding the market. In addition to expanding the market, we believe we grew our relative share of this market in Q4. In fact, just recently, we received a significant order from a major Genome center that is not previously scaled on Illumina technology. Part of this market expansion also includes whole human Genome sequencing through the Illumina Genome network. Our partner-based model allows us to flexibly leverage sequencing capacity across the network in addition to our internal capacity. We recently received a multi-hundred Genome order from a major pharmaceutical company through this service, and have over 1,000 Genomes in backlog from a wide range of customer types. Early in January at the JPMorgan Conference, we shared several recent improvements that we've made to our TruSeq SBS chemistry. One result of these improvements was a demonstration of multiple sequence runs, which have generated over 1 terabase of data per run or 1 trillion basis of genetic information. By our estimates, we believe that one run of HiSeq at these levels generates more sequence data than the entire world generated in 2007. With these system improvements, we've been able to increase throughput a factor of 1,000 from 1G to over 1T of data per run in just four years. These 1T runs took approximately 14 days and utilized 2 x 150 base paired-end reads. We plan to commercialize the underlying improvements through a series of software and reagent releases beginning this spring. The first of these is a 2 x 100 base pair kit, which is currently running at early access customer sites. We expect this kit to enable the entire HiSeq install base to generate about 600 G of throughput per run with no increase in current run times and no upgrades to the hardware. Through new entymology in these kits, we've been able to increase cluster density while simultaneously improving accuracy. Customers will now be able to sequence five Genomes per run, which will drive down the reaging cost per Genome well below $5,000. Much of our development work in sequencing has been focused on increasing throughput. As we've done that, we've also brought down chemistry cycle time to keep overall run times in check. We've recently made major improvements that have allowed us to bring our chemistry cycle times down by a factor of five. And last month, we debuted the benefits of this breakthrough in our new MySeq [ph] personal sequencing system. MySeq is an easy-to-use, low-cost sequencing system with a compact footprint, just 2 feet square. The beauty of MySeq is that it utilizes the same TruSeq SBS chemistry that powers the HiSeq and the Genome Analyzer, and supports complete work flows from sample to data in less than eight hours. The sequencing capabilities of the platform are entirely flexible and can perform single, 36 base pair reads, all the way up to 2 x 150 base pair reads, yielding as much as 1.5G of data depending on the application. This represents more than 10x the data generated by comparably priced systems. HiSeq will be priced below $125,000 all in. An additional breakthrough of MySeq is that it does not require any peripherals to support sample prep or analysis. The cluster generation step of the TruSeq workflow is performed on the system and eliminates the need for a cBOT instrument. MySeq has an onboard computer powerful enough to handle control of the instrument, base calling, and secondary analysis. MySeq is also capable of cloud connectivity so sequence data and analytical reports can be uploaded and shared easily between teams. We took the human interaction concept pioneered on the HiSeq platform to the next level on MySeq to facilitate broad market adoption. The reagents for MySeq are premixed and packaged into a simple, single reagent cartridge that includes an RFID [ph] chip for tracking. The customer simply puts the sample into the cartridge and places it in the system, looks to sell, which is also RFID-enabled and they're ready to sequence. We expect total per run prices to range from about $400 to $750 per run. Along with the introduction of MySeq, we also announced the acquisition of Madison, Wisconsin based Epicentre Biotechnologies. Epicentre is the leading innovator in next-generation, sequencing sample prep technology, has also developed a broad portfolio of specialty enzymes. I'm happy to welcome the entire Epicentre team to Illumina and look forward to great contributions to our product evolution. The leading product in the Epicentre portfolio is the Nextera assay that enables a radical reduction in library prep time across all of our sequencers. In the eight-hour MySeq workflow I mentioned previously, cycle time reduction is a key driver but the new Nextera library kit is as well. The Nextera kit allows ultralow sample input and can generate sequencer-ready libraries in less than two hours. We're extremely excited about the future possibilities that we see with Epicentre to further improve and simplify the workflow across our product lines. The initial feedback we received on MySeq has been extremely positive. Last week, at HEBT, customers commented on numerous applications where they believe MySeq will offer distinct advantages, such as targeted resequencing, small sequencing, clinical screening applications, and QCing quantification for high throughput labs. Interestingly, several of our larger customers commented on how they will use the system to rapidly and cost-effectively develop new applications, which can then be ported over to the high-throughput HiSeq system. We will begin taking orders for MySeq in April and plan to ship early access units by the middle of the year with volume production in Q4. In many respects, Q4 was our best quarter of 2010. We generated very strong year-over-year and sequential revenue growth. We successfully transitioned a significant number of customers to the HiSeq platform, and we're able to maintain gross margins within our long-term model. Through disciplined expense management, we delivered operating margins over 31% and generated a record $81 million of operating cash flow resulting in free cash flow of $69 million or $0.49 per share. Our Array business is stable and continues to demonstrate a valuable and complementary relationship to our sequencing portfolio across multiple markets. Together, the two technologies are a powerful combination unique to Illumina and validated by the strong uptake of the HiScanSQ platform. We continue to believe that rare variant based discovery will renew GWAS growth in the back half of the year as we see the results of ongoing proof of principle studies and launch the Omni 5. Looking back on 2010, it was a transformational year for the company. We grew total revenue by 35% and generated $223 million in free cash flow or $1.66 per share. We managed through the launch of the most successful product in the company's history and generated over $1 billion in orders, ending the year with a record backlog of approximately $300 million. In 2011, we see no signs that the demand for sequencing will slow down. The global funding environment for our markets remains stable, the continued improvements in TruSeq chemistry will enable whole Genome cost on HiSeq below $5,000 opening new markets and enabling much larger sequencing studies. And finally, the introduction of MySeq has rounded out our portfolio very nicely to deliver TruSeq chemistry and next-generation sequencing to a broader market, including exciting clinical applications. Thank you for your time, and we'll now open the lines to your questions.

[Operator Instructions] Your first question comes from the line of Marshall Urist with Morgan Stanley.

Jay, if you might talk about how you're kind of thinking about the MySeq market opportunity, both kind of relative the relative side of the opportunity kind of selling to the current installed base and then how you see starting to transition to new customers be it either on the academic side or clinical and maybe how the timing of those two markets might play out?

Yes, I certainly think the low hanging fruit will be to sell it to existing installed base. These are customers that we call on every day and are already familiar with the SBS chemistry and the way our data is analyzed. And we think in the high-throughput labs, they'll be using MySeq for doing incoming quality control. They will be using it for application development and in many cases, as a way to deploy targeted application either for targeted sequencing, amplicon [ph] sequencing or emerging clinical applications. I think in parallel, we're going to begin selling to the broader market as well. The easiest orders of course will come from installed customers, but we think there is a very ripe and large opportunity to begin to go after the remaining portions of the CE capillary market. Not all of those are going to be accessible because some of them are very sticky markets like forensics and some of the food and water testing markets. But certainly, much of that market that's used for doing amplicon resequencing as an example will be accessible very quickly to the MySeq technology and be able to be done at much lower price. So I think we're going to open up that academic base quite broad and very quickly. We think in the long run, clinical applications are going to become very important in MySeq and the consistency of the SBS chemistry is important there, the accuracy of the data, ease-of-use, all of the factors we talked about are important. And the feedback we've gotten already is that this system really hits the sweet spot of performance in throughput the clinical customers are looking for. Obviously it will take a long time, some number of years to develop a large number of applications in the clinic, but this will enable that application to start right away.

Christian, one follow-up on the gross margin guidance, I just wanted to clarify, is that 70% a GAAP number? And then if we think about the cadence over the year, where should we be thinking about exiting the year from a gross margin perspective as the trade-ins are done and then as you start to get more pull through on the consumable site for HiSeq?

Marshall, the 70% is a non-GAAP number but if you look at our stock comp expense and other expenses that fall into that gross margin line, they're really insignificant. It's really the stock comp expense and the intangibles. So it's likely we would be above it both on a GAAP and a non-GAAP basis. As far as the cadence goes, it's really going to be, as we said in the first quarter, we have some more of that HiSeq trade-in or GA2X trade-in programs. We have some of those units to get through. So that will have some impact in the first quarter. And then as you get into the year, the installed base is becoming big enough on HiSeq as we said in terms of the consumable pull through. You should really start to see the consumables continue to move, and that will help the gross margin as you get into the second, third and fourth quarter.

Is that going to be ratable over the year? So we could be exiting the year on a sort of on a low-70s number?

We haven't made any comments with respect to that. I think at this point, we're just talking about how do we get to 70% for the year and then obviously the first half is lower, the other half is going to be higher. But we're not ready to talk about any specific quarter yet.

Your next question comes from the line of Ross Muken with Deutsche Bank.

As we think about the AGBT meeting, obviously, lots of posters, lots of presentation and lots of great buzz on the HiSeq. If you had to sort of distill from a scientific perspective or from an advancement on the box perspective a few things you kind of walked away from and thought, wow that was sort of a powerful kind of example of what the technology can do, and how eventually this is going to be used in kind of more practical applications. Is there anything, one or two things you can point us to that you would say, for you that sort of stood out?

It's Christian, I'll start. One of the things that struck me was when David Bentley, our Chief Scientist, was giving a presentation, and it was in one of the sessions with all of the different companies. And what really struck me was a collaboration. He was talking about collaborations and cancer where we've taken the technology as far enough along now where we're really working on the science and the clinical applications of the technology where some others are really still focused on the technology itself. And in particular, David presented some very interesting work in cancer and I think, it just kind of shows where all of this is really going in terms of developing systems that are easy enough for people to use to get really meaningful clinical data and make discoveries that are going to affect patients. To me, that was the most interesting part of the conference. Jay, would you have anything to add there?

I guess from a technology perspective, what struck me was how routine it is to sequence complete human genomes now and to do all [indiscernible]. I think in the next couple of years, the [indiscernible] market is going to be growing quite quickly. And our kit is enabling to that from a price and performance perspective. But if you look back just a couple of years at how difficult it was in sequencing a human genome, if you compare that now, it's been a radical shift and people are talking about doing very large collections of human genomes, and not worrying about that. The advancement obviously are huge across the platform technologies. I think you've seen great advancements in sample preps like what we talked about with Nextera. And obviously our competitors are making their sense of advancements. I think the other thing that stand out is the challenge that we all have to face in the biokinetics. We may and others are focused on and working hard with collaborators to figure out how to speed up the back end data analysis and how to do a better job of extracting the biological relevance out of these very large data sets they are operating.

And maybe Jay, you talked about a lot of applications for the MySeq and potential areas where you guys can you address, it's a bit of a different customer base and different sales approach than some of the traditional research markets with some of these new applied applications. How do you think about building the commercial organization on the sales and support site and how is it sort of different than maybe some of the other challenges that you faced with the existing business that you build out over the last five years?

I think we're going to continue to grow the sales force, of course. That would've happened with or without MySeq in some ways. I think you'd be surprised at how similar the customer base is. Our customers have been pushing us very hard in the last couple of years into this clinical research market, and that's on our array technologies, in the study of genetics. We have so many customers using HiSeqs now to do research into clinical applications, and what we've done now is provided the platform to deploy those applications very cost effectively and very broadly. So I do think that the initial customer base is not that different. Now if you look down the road a few years, we may be selling much more to hospitals and markets like that who are actually deploying these clinical applications in a large way. And I think we'd love to do some evolution in the sales force of the sales force as we get towards those markets.

I think, Ross, one of the things that was really interesting was really, and we said this in the prepared comments, was how interested our large customers were in MySeq because it's such a rapid turnaround system and because it's got the same chemistry with the same high accuracy that they can develop their own applications, much faster than they can ever do on the HiSeq system. That actually surprised me a little bit of how excited they were about the product.

Your next question comes from the line of Tycho Peterson with JPMorgan. Tycho Peterson - JP Morgan Chase & Co: Jay, I just want to make sure I understood your comments on the Array business. I think you talked a little bit about strong orders, but then you talked about customers taking time to analyze rare variant, work off the new Omni 2.5. How do you know about the cadence of array sales at the start this year? You have the Omni 5 later on. I guess are you a little bit more conservative in the first half of the year for arrays, and then it picks up in the back of there?

I think overall that's probably right. What we're seeing is the market subdivide in some segments in a couple of different ways. We're seeing the camp of researchers that believe strongly in the hypotheses. They're the ones who are buying the 2.5 and they are now immersed in doing these initial proof of principle studies. And then we have a lot of people on the sidelines sort of waiting to see what happens out of those studies. So I think there's going to be as we get into the middle of the year Q3 sort of an inflection point here based on what happens in these initial studies. So that part of the market, I think will start growing in the back half presuming the kind of data come out of projects that we and others expect. Then you have the camp of researchers who are still very focused on the more common variant hypothesis. And what we're seeing there is a greater use of the Omni Express than we might have thought, continued use of Omni Express but with very large uses of sample sizes. And the goal there is to get greater statistical power out of their studies by increasing the sample size or in some cases, doing secondary populations from the ones that have been previously and so, there's high sample numbers there but of course, those chips are lower ASPs than what we have in the higher end on the 2.5 category. Then and of course, we have the third segment of our Array business which is focused and custom which continues to grow very, very nicely. I think if you add all those segments together, we're probably more conservative in the front half of the year presuming things go the way we expect, stronger in the back half. Tycho Peterson - JP Morgan Chase & Co: As we think about the performance enhancements on the HiSeq, obviously great, great throughput getting to a terabase. I guess in terms of the steps you laid out going to 2 x 150 read lengths and things like that, would it be a serious of through the course of the year after you introduce the 600 gig improvements or will it be kind of batch products released later in the year gets customers ultimately to a terabase?

Well we group these together. Because customers don't want to make too many changes all at one-time, and often these production shots need to be validate any particular change we make. So we do try to couple them together. And if you think about it, the 600G performance level is essentially the same as one 1T in many ways because the 1T was a 2 X 150 run, and the kit we're going to release is the 2 X 100 kit. So you have essentially a 50% difference just made by the reblank. And our best judgment based on how the customers are actually using the systems in the field. Is that most customers don't want to go up to 150 basis and extend the run times. And so, the kit we're going to release is that 2 x 100 kit, which gets some roughly 600G number. And with that, they have most of the same capably that the certificate to the 1T if you think beyond that, we have lots of other things and lots of dimensions and scalability remaining in our system across all the fronts that we've moved before. Software, better image analysis, higher cluster density and faster cycle times, larger flow cells, all the things that we've done in the past, better etymology are all still in front of us. And as I said in the JPMorgan Conference, we don't see the end of the runway for SBS chemistry anytime soon.

One other thing to point out on the chemistry that's going to be launched 600G kit is the dramatic improvement in reduction in bias [ph]. I think that's actually, not only do we get increased cluster density but we also get better lower error rates and significant reduction in bias. So using this kit for potentially covering genomes with a little bit less coverage is also a possibility down the road. Tycho Peterson - JP Morgan Chase & Co: On Epicentre, can you just talk about the opportunity to leverage that business across the existing portfolio today? And ultimately, is there an opportunity to kind of on-board that sample prep onto the platform as well?

There's many ways we're going to use the basic technology that Epicentre brings. The obvious first thing to do is this Nextera kit, which works beautifully with library prep. We'll deploy that with MySeq, but it of course works across a whole range of our sequencing product line. And in fact, many of our existing HiSeq customers have already been using this Nextera kit. We're going to continue to evolve that kit and there will be multiple versions of it. There was a poster at AGBT about a PCR free version of the Nextera kit, which will be an even better performing kit in terms of sample input and freeing up from any biases that coming from the PCR reaction. In addition to that, the core enzymes that Nextera has can be incorporated in other places in our product line. In fact, we're currently a licensee of several of their other enzymes that we use in the Array business. So obviously, now we will have the ability to use those enzymes in cost rather than price. And we have enzymes that we currently buy from other third parties that we think we can begin to transition over to using enzymes that Epicentre would provide. So we have sort of a broad way to utilize the set of technologies. And then we'll take our development team and really focus on the sample prep methods of the future. On the boarding the sample prep, there's always tradeoffs in doing that. It's technically possible, but it's something that we'll have to look at and study in terms of how we map out the product roadmap. Tycho Peterson - JP Morgan Chase & Co: On HiSeq versus MiSeq, do you envision any customers evaluating between the two platforms? In other words, are customers going to be looking at MiSeq and ultimately saying,"We have higher throughput needs and will naturally migrate to HiSeq?"

I don't think in a serious way. I mean, somebody who didn't know the product line at all may take a look across the board at all the products, but they're very different in terms of the usage of the system. One is a high-throughput production system and the other is good for targeted work and smaller genomes and clinical app. So I don't think we really have anybody making a trade-off decision.

Your next question comes from the line of Isaac Ro with Goldman Sachs.

Just first off on informatics, I know that the key pain point that's emerging [ph] for customers. So I'd be interested in knowing how you think your approach there would differ between the HiSeq versus MiSeq customer base going forward?

Well, it won't really be very different at all. I mean, these systems are intended to work hand in gloves, and the chemistry is the same, the data format is the same, the informatics channel is the same. We do think that the market that MiSeq will sell to will probably be one that would want to do less computing onsite. And so we're working to make sure that MiSeq is cloud-enabled so that you don't have to have any computers to run a MiSeq whatsoever. And the data can be moved up to the cloud and then the compute done up in the cloud. Whatever we do to implement that on MiSeq would also apply to HiSeq, so the implementation wouldn't be different. It's just that the customer base for MiSeq would probably gravitate more towards that solution than having in-house compute infrastructures.

And then just secondly on just sort of philosophically as we think about the various technologies that are out there, given the big decreases in costs that you guys are enabling on HiSeq and therefore, the ability to do some really deep coverage. Could you maybe give us a sense of the percentage of sequencing applications that you think really necessitate long leads exclusively. And sort of a result, how important is that capability down the road for you?

By long reads, you mean an excess of 1,000 basis, that kind of long reads?

That's the right number. Yes.

We think there's some applications where long reads can be helpful. But they're actually few and far between, and we think it's a pretty small segment in the market. The vast majority of applications where that's necessary, we can get at using our paired-end technology. Currently, we support paired 150 base reads, and the utility of that paired-end technology increases as we have the ability to put longer inserts in. So we're continuing to work, and we'll be working with Nextera to figure out ways of putting longer inserts between the fragments that we read. And that will take even a greater share of the potential long read market. Probably the one area that we're working on importantly to try to get at is this idea of long-range HLA [ph] typing. And so we want to get at that as well. But we think the current technology we have is going to get at most all of what happen a long leads, and then there will be uniquely very few applications that we can't do.

As you think about all the new instrumentation platforms you guys have put out over the last year and a half, you have a pretty wide range of offerings at every price point, is there any white space up on the board at this point for implementation offerings that you think you could offer the market that they need that they don't have to take the.

We think we have most of the market covered with what we have today. Obviously, on our drawing boards, we have a five-year evolution of all these technologies and how they may go and migrate in the future. And obviously technology like Oxford Nanapore [ph] is a component of that. We have multiple internal development programs both in research and product development that are looking at fundamental ways to improve sequencing, and that has to do with everything from the chemistry to the hardware to the imaging to the sample prep to the bioinformatics.

Your next question comes from the line of Amanda Murphy with William Blair. Amanda Murphy - William Blair & Company L.L.C.: If you think about your focus on expanding the sequencing market beyond kind of the current and next-gen installed base, could you just talk to the opportunity from a HiSeq standpoint? You talked about the MiSeq a little bit, but just curious how users might leverage the HiSeq that don't currently use platforms today?

Leverage them for what? Amanda Murphy - William Blair & Company L.L.C.: How users that sort of don't have an instrument today might utilize the HiSeq?

I mean, if you look at our installed base of HiSeqs, the vast majority of customers own one. And so what that points to is the fact that we have been able to expand the market greatly with, first, the Genome Analyzer and now HiSeq. And we think we're going to be able to continue to do that for research groups who can find the funds because of what HiSeq allows them to do such phenomenal science that was virtually impossible just two years ago. And so we think the market will continue to expand in that dimension. And clearly, we think that the customers that already own one were going to continue to buy their second, third, fourth and fifth. So as we plot out the demand curve for HiSeq, we don't see any reduction in the sort of baseline demand. Now clearly, if you look at 2010, there were two components to the demand side for HiSeq, there was the upgrade portion, which is not going to be recurring, and then there's the steady state portion, which is sort of the organic market growth. And we don't see any slowdown in that organic market growth whatsoever as we look forward. Amanda Murphy - William Blair & Company L.L.C.: And then just another one on the HiSeq, in terms of the consumable or the reagents of usage, I think you mentioned it was kind of trending around $350,000 annually. Just curious how to think about that over the next year. And are any of the upgrades that you're going to roll out, did they impact that number at all?

They don't impact it materially. So we'll continue to monitor that number. This is the first quarter where we have enough instruments in the field to be able to get a handle on the number initially. And it's likely to bounce around a little bit for the next few quarters, as we increase the installed base that's actually utilized in production. And it's a little too early to forecast where it could go. So we're going to stick with the $350,000 to $400,000 number for now until we have better information.

Your next question comes from the line of Doug Schenkel with Cowen and Company.

As you guys have talked about a bunch over the last several quarters, you've really scaled up your HiSeq manufacturing. I think again you said today by a factor of 10 over the last year. Presumably this is going to allow you to work through backlog over the next few quarters. Is the broader thinking here that you're going to work through the backlog sufficiently enough to address demand and then shift production capacity to MiSeq? Or are you thinking at this point you might actually have to increase capacity again to address both parts of the business?

I think you hit the nail on the head there, Doug. I mean, the plan is that we will -- I think we're sort of at a peak of HiSeq production now, and we'll be operating at this level for another quarter or two, bringing down the backlog, and then we'll begin to transition both facilities and our employees over to manufacturing MiSeqs. MiSeq is going to be much easier to manufacture because it's a vastly simpler instrument and the test time is one day rather than seven days. So we're going to be able to make those in much higher volumes in smaller space with fewer people. But Christian has laid out this plan out beautifully, where we have this ability to -- we have an overshot on HiSeq because we can take that capacity and been around MiSeq.

Doug, we just opened up basically a new space up in Hayward, so where we can manufacture HiSeqs and MiSeqs and also, the HiScan and HighScanSQ side-by-side. So we can really leverage our people resources. One of the key things that Jay continues to push on us on operations is commonality of components and this and that. So we have a lot of commonality at least between the HiScanSQ and HiSeq system. So we get leverage out of that as well. But there's still going to be very high production levels of the HiSeq system for the next several quarters. So we're still going to be busy, that's for sure.

Just to be clear on that, you're clearly not factoring in a dropoff in utilization, given what you talked about in terms of how you're going to track the gross margins this year? Is that correct?

No, I think if you think about overhead utilization and things like that, I think we're going to be very, very well-utilized, if not, overutilized. We'll probably use a combination of overtime and multiple shifts to try to make sure we make the demand. One of the reasons why our goal is to be shipping the MiSeq system at volume in the fourth quarter, which means we have to be capable of producing reasonable numbers of it in the third quarter as well. So the operations team has a lot of work cut out for them in the first half of the year here to get ready for that.

And maybe sticking with gross margins for one more, I think you guys talked about a target, three-year or five-year gross margin range of 65% to 70% at last year's analyst meeting and your guidance for this year is to get up to 70%. Is there any reason to think that you can't bust through the high end of this guidance, as the consumables continue to ramp on these new instruments?

There's certainly lots of factors that could push us above 70%. If you looked at formulas for reagent growth and the margins we get on reagents, what we expect to happen in the mix in 2011 and 2012. But there's factors that drive it the other way as well. And so we are always leery about what could happen in -- Array overhead is one area that we were very conscious of all the time because as we continue to put more samples on a chip, our Array efficiency is just unbelievable. And so we needed to be able to continue to absorb that overhead. And so we moderate all those downside factors as well. And it's the combination of those things that resulted in the guidance that we gave for the year.

But I do think we're definitely pushing to maximize our opportunity in terms of growing the gross margin. Another reason why we try to be a little bit conservative here is also for our internal programs because we want -- really gross margin drives how we think about all of our spend programs and growth associated with our expenses. And so we kind of look at a whole consolidated -- a holistic model, so to speak, of what the business should look like. But as Jay pointed out, there's factors on both sides of the ledger that could push us well above it or be in line with it.

Your guidance seems to incorporate the assumption that you get some pretty good leverage at the operating line far beyond the gross margin improvements. Any chance you'd say whether you expect to get more leverage in the R&D or SG&A this year?

Well, I think the percentages you saw us deliver in Q4 were pretty darn good. And so I don't think we're going to set a goal to do materially better than what those percentages are.

Our next question comes from the line of Jonathan Groberg with Macquarie.

If you would be willing to characterize how you feel your commercial capabilities right now are set up in some of the more emerging geographies around the world? And maybe what you think you do or don't need to do in some of those geographies? And then the second question is related to GWAS. I guess, given the delays that you seem to be seeing with respect other demand or developments like the 5M, when do you maybe just stop investing in GWAS technology and focus your R&D dollars in other areas of the Array market that maybe will grow more, and particularly you see the development that's happening so quickly with sequencing as you just alluded to with what's happening with HiSeq?

First, on the commercial side of the ledger, we have all of the world's materially large markets covered very well with our commercial team. We have quite a number of people on the ground in China now. So we think we have the vast majority of what we need to be covering of the key accounts in China covered. The other large geography where we're not direct is in India, but to be honest, with our kinds of products, we think that's a pretty small market, and so we deal with India through a distributor. And the most recent addition we've made there is opening our office in Latin America, and that's a small sales office really focused on three countries, Brazil Argentina and Mexico. And we think that takes the ball of what we're going to be selling in South America in our types of products at least. I guess the other segment of the market that need to lower price distribution is the Eco product. And that one in international market, at least, we're dealing largely through distributors because that is a market segment where our current sales force doesn't largely call on that set of customers and we need more feet on the street. And so in that segment, we'll use distributors. With respect to GWAS, we're in a bit of a -- it's a very high-end here in the 2.5M and the 5M. We are waiting to see what happens with these proof of principle studies. As we get to the back half of the year, I think depending upon what we see, we may alter our investments in R&D. And I would say that we, in fact, have been doing that over the past few years. If you look at what our investments were in 2007 or 2008 in sequencing versus Arrays, the investment equation has swung pretty substantially in our overall R&D resources towards sequencing. And I think that will probably continue. We do have lots of great ideas in ways to continue to advance the Array market, but we may or may not need to implement those on a fast time frame.

Your next question comes from the line of Quintin Lai with Robert W. Baird. Quintin Lai - Robert W. Baird & Co. Incorporated: So you guys finished 2010 really strongly, sequencing, hitting on all cylinders. But as I look at the 2011 guidance for 20% top line, maybe could you dive a little bit deeper? And so what part of the business is -- is it tougher comps as you said, Jay, kind of the upgrade cycle? Or is it just an instrument tougher comp coming up?

Our top line guidance, Quintin, is clearly of a much larger number in this year than what we had last year. It's an equivalent guidance we gave at the top line coming into 2010. And we always take into account a wide variety of factors, having to do with our new product launches, their timeliness, what we think the competitors might do. And we do have a bit of caution on the Array part of the business coming into the year, and we want to make sure that we're adequately cautious with respect to what's going on in the Array business. We do continue to be very bullish on sequencing, no doubt about that. But it is a competitive space, and we have lots of companies out there that are trying to grab pieces of the sequencing market. Quintin Lai - Robert W. Baird & Co. Incorporated: So kind of going to the microarray, you had a really good instrument quarter, it sounded like in the fourth quarter. Was that just upgrades to existing, or are you seeing new entrants as well?

Well, it's new instruments and very strong demand for HiScanSQ. We had built up a bit of a backlog on HiScanSQ, as I mentioned in my prepared remarks, because it uses common components with HiSeq. We were biasing most of the shipments toward HiSeqs because they have higher reagent assumption. And so we were building up more of a backlog of HiScanSQs and we were able to ship out a material part of that backlog during the fourth quarter. The incremental order was still strong, so we didn't exhaust the backlog but that was certainly one factor in what happened in every instruments. Quintin Lai - Robert W. Baird & Co. Incorporated: The tax rate, I know that you moved some microarray manufacturing into Singapore. Given the strength in sequencing, kind of dovetailing off of the R&D towards sequencing, any thoughts of more tax strategies to moving maybe some manufacturing to that area?

Quintin, this is Christian. So in the fourth quarter, you saw our tax rate was improved a little bit. But quite frankly, it should have been even lower because of the passage of the U.S. R&D Tax Credit, the retroactive adjustment. But we took advantage of that opportunity by transferring some sequencing-related intellectual properties in Singapore so that will set us up for a lower tax rate in the future. So that's on the tax rate side on the structuring. So now we've also set our corporate goals to be manufacturing a significant amount of our sequencing consumables in Singapore by year end this year. And so jay is going to hold me accountable to that, I think. We're going to work really diligently to move sequencing consumable production out of San Diego, or portions of it out of San Diego at least, and expand into Singapore. We just took down some more space than there, and actually the space will be built out in a couple of months. And we'll start qualifying reagents. And hopefully by the back half of the year, we'll be able to report that we're producing a significant amount of our sequencing consumables out of Singapore and that's exactly right, Quintin. That's going to help our tax rate. It will probably take a year or two for us to get the full benefit of it, but activities are starting this year.

The first time to move instruments over there, so our easier to manufacture and more mature instruments are moving over Eco because it is relatively easy to manufacture and be expressed because it's a relatively mature product.

Your next question comes from the line of Dan Leonard with Leerink Swann.

So on your trade-in program for the GA, it sounds like you're concluding your trade-in offers at the end of the first quarter. Is that the correct read that you will no longer offer any compensation for trade-ins past Q1 2011?

No, that's not quite right. We will be offering, because we think it's really in our best interest and the customers' best interest to get them to HiScans if they can get the money, that we will continue to offer some incremental discounts for customers to convert over the HiScans. Again, only on a one-for-one basis if they trade in the Genome Analyzer. What has gone away are the deep discounts that we put in place for customers that had just purchased Genome Analyzers. And you recall, we launched this in January of last year. And the deepest discount was for customers who had just taken the system maybe a month before. And so they effectively got a full trade-in credit for their Genome Analyzer. And then we extended that back through customers that had taken Genome Analyzers in the third quarter of 2009. So it's really those two quarters where we had the deep discounts because of the unique situation, where we had to treat those customers well to help migrate them to the new technology because they had just bought the prior technology.

Jay, how do you view the funding for GWAS 2.0 if those pilot studies are successful? Is that work going to compete with sequencing for funds or are these separate funding buckets in your mind?

There will be some competition for funds, but we don't think that much. There still is a fair amount of money that's out there and on-hold, waiting to see what happens in GWAS. So we do think there will be bolus of funds that will come in that direction if these studies prove out to be extremely useful. And obviously, we're confident in that, but we have to wait and see the data.

Your next question comes from the line of Bill Quirk with Piper Jaffray.

Jay, in terms of new order from the specific large Genome Center, did that, by chance, represent the majority of the Genome Center orders in the quarter?

Yes, the order was received after the quarter.

Secondly, just thinking about everything that's going on at AGBT, can you talk a little bit about customer reception to the TruSeq Exome capture products? Obviously, that was a pretty significant topic of discussion, at least I'd argue it was, over the past week as well.

Yes. I think the reception has been fantastic. It's being tested quite broadly now by many customers. Some of the large Genome Centers have tried it. It offers big advantages, we think, in cost certainly and in performance that's at least as good or better than the existing kits on the market. And we're also going to sort of morph that technology into the custom version of that kit. So you'll be able to use very similar kinds of workflows with custom-targeted products that we're going to launch right at the time that we launch MiSeq. Those customers sort of buy into the collection of products that we have, they'll be able to do fully custom if they want or all exome or whole genome, and they'll be able to use some of the Nextera technologies throughout that pipeline of products.

Your next question comes from the line of Derik De Bruin with UBS.

You mentioned that you took a number of whole genome orders from some customers. I'm just wondering, you've got a backlog. What's the ASP right now for a whole genome sequence? And then on a follow-up on that one, I want to ask the questions, like there's a lot of debate at AGBT about whole genome analysis versus the continued use of all exome analysis. What do you think if the tripping point in terms of pricing to get more people to do whole genomes versus exomes?

We don't give out the specific pricing because it varies based on volume in a particular situation in the account and what else we're doing with that account. But I can say, in almost all cases, the prices are $10,000 or somewhat lower than that. And obviously, those prices go down with volume on the order. I certainly think as whole genome sequencing gets down to the $2,000 range, you're going to start having fewer all exomes done. Some people have already wanted to do all exomes and they've moved to complete genomes today. I don't think the majority of people have done that. But as you get down to a couple thousand dollars for a whole genome, maybe half the market will be whole genome and half will be exome.

So you've got all these GAs you've taken as trade-ins. Where are they going? And what's the economics on the resales?

Right now, Derik, what we're doing on the GAs is as we bring them in, we refurbish them and we're moving them into markets, like we talked about we're opening up Latin America, for example. So the GAs are still delivering 95g of high-quality data. And so in some markets that are extremely price-sensitive, we're moving some of those instruments there. We're leveraging some of those instruments in various kinds of collaborations. And then we're also -- some of the instruments, quite frankly, we will use for spare parts and kind of to support our service programs over time. So we don't expect to resell every Genome Analyzer that comes back in the door. But in 2011, we expect to move a pretty reasonable number of these back out into the field to be generating a consumable pull. The cost basis on the first bolus of these isn't zero but it's low. So it gives us a lot of flexibility in terms of how we market them. It opens up the opportunities for doing creative deals for particular kinds of customers or towards things like reagent rentals, which we might not otherwise do when the capital equipment is expensive.

Could you just give us a little bit of clarity on amount of your sales that are split between sequencing and arrays?

We haven't really broken that out. I guess, all we could can say now is that sequencing is bigger than Arrays.

Your next question comes from the line of Nandita Koshal with Barclays Capital.

My first question is what is the thinking on regulatory approval for clinical use of the HiSeq and MiSeq? And have you started conversations with the FDA?

We've had broad conversations with the FDA, and our customers are moving us, asking us really across the product line to begin to push our products toward achieving regulatory approval. And we're certainly seeing this in the Array business, driven largely by the side of genetics market. And so we're worked on the iScan to get it approved as the Array scanner. We've already done BeadXpress. And we're right now evaluating what we do first in sequencing. Clearly, MiSeq is a product that will wind up being in the sweet spot of clinical applications. I think that's a no-brainer. We're also looking at doing HiSeq, and we'll probably make the final decision on how and when we would do that in the next month or so.

And then where are we in terms of shipment from the BGI order? And do you know how the consumables pullthrough might ramp up on that? The reason I ask this is I was wondering if the consumables manufacturing capacity and the quality control resources that you need are enough to satisfy the demand from sort of this expanded HiSeq installed base that's coming online?

Yes, this is Christian. We have probably about 30% left to ship under the BGI order. But if you remember, the revenue recognition is over a period of three years, and it's kind of a ratable recognition process. They continue to scale up and operate at full capacity, and we've been able to keep up with them. Although at times, it's been challenging, but we have been able to keep up with their demand, and I don't see any roadblocks to keeping up with them as we install the rest of these systems.

Yes. I think on the reagent side, Christian and the operations team has done a good job of scaling our reagents. And we're continuing to meet demand, with few exceptions where we sometimes have a hiccup here and there as you do when you're scaling rapidly. But overall, I think we're doing pretty well there.

Your next question comes from the line of Amit Bhalla with Citi.

Question on just in terms of MiSeq and the supply side of the raw materials, can you talk a minute about how they're similar or differences between MiSeq and HiSeq on the supply side?

There's actually not a lot of commonality of components between MiSeq and HiSeq. MiSeq was a ground-up redesign targeted for a different market, much simpler implementation. It's a single channel in the flow cell, so the way actually the motion work is totally different, fluidics are totally different and the optics are totally different. So there's very few parts that are going to be common between HiSeq and MiSeq.

So along that line and then, can you just expand in terms of the number of suppliers you have on the MiSeq side? And is that all squared away and right in time for a 4Q big scaleup?

Yes. I don't exactly how many suppliers offhand, but most of the key elements of the system have already been flushed out in terms of the suppliers and supply chain. As I said earlier in the call, it's going to be -- in order for us to manufacture at significant scale in Q4, we have to be operating at pretty reasonable scale in Q3, which means that we're in the process of scaling up a lot of these suppliers between now and, say, June. So we learned a lot on the scale up of the HiSeq system. I think the demand exceeded our expectations. One of the benefits of having announced the product earlier than starting commercial shipment is that we can actually anticipate the demand a little bit better, scale our vendors, work on our quality control systems, et cetera. Jay also pointed out a very, very important part of this is that the QC time frame or the FIT -- our functional testing is only a day so we can rapidly turn these systems around. And the engineers have done an amazing job of creating a system that's very manufacturable. So from a scale-up perspective, we really are focused on the supply chain, the operations itself and then making sure that each system goes out at the appropriate level of quality. So a lot of work to do, but we're looking forward to tackling this challenge.

Christian, what were the stimulus revenues in the quarter?

We didn't disclose that actually.

Your next question comes from the line of Zarak Khurshid with Wedbush Securities.

I just had a question on Epicentre. Any detail as to what you paid for it? What is it doing in revenue? And are you going to be selling the technology to third parties going forward?

We haven't discussed any of the financial elements of the transaction. They have a very broad product line of products that we'll break into two categories. One category will be those that work directly with Illumina systems and those we'll largely rebrand in Illumina packaging and Illumina colors. But then we'll continue to sell Epicentre products under the Epicentre brand, and we expect to continue to do that for the most part. I mean, we're certainly going through a portfolio review to look broadly across all those products and make sure they all make sense to continue because some of them may be enough in revenue to continue to market for us. But we'll have an evaluation done in the next couple of months. And clearly, from an R&D perspective, the forward going focus is going to be on reagents that work with Illumina systems.

I was just curious, you mentioned the all exome sequencing opportunity and how that's growing. Could you talk a little more broadly about RNA-Seq, ChIP-Seq, any of the mid-tier [ph] submarkets and how large those are and how fast they're growing relative to sequencing and resequencing?

Most of our customers run multiple types of applications. So it's sometimes a bit hard for us to pull it apart. But the best estimate is that about half of our customers run RNA-based applications. And some of those also run sequencing applications. So it's a pretty broad usage in the RNA space. And I think you're beginning to see an acceleration of the transition from expression-based arrays on the sequencing because of the cost and the quality of the data and all the reasons that we've spoken about before. So we continue to believe that the RNA side is a fantastic opportunity. And in fact, we haven't mentioned this directly but Epicentre has some really good kits that work with RNA as well. And those will become Illumina-branded and sold with our RNA sequencing portfolio.

Your next question comes from the line of Sung Ji Nam with Gleacher & company. Sung Ji Nam - Gleacher & Company, Inc.: First one, in terms of your expectations for HiSeq 1000 versus 2000, do you anticipate meaningful demand for 1000? And is there really a clear, I guess, segmentation in terms of the market there?

I mean, the market segmentation between those products is largely driven by funding. And so our view is that anybody who can afford a HiSeq 2000 would buy that as opposed to a 1000. But there are certain customers who can't get enough money, maybe you're putting up money from a couple of groups. And because the 1000 is upgradable to 2000, if they can only get enough money to buy it, it's a great way for them to get into the pipeline of our products and to not constrain themselves in the future because they can always do the upgrade. So that's really the market for the 1000. We don't think it's distinct in terms of having a separate throughput category that somehow is different from a 2000. Sung Ji Nam - Gleacher & Company, Inc.: And then in terms of given your current portfolio of products including MiSeq now, do you anticipate your strategy for BeadXpress to change going forward in terms of your partnerships and things like that?

No, BeadXpress is really targeted for different kinds of applications. There, we do proteins and genotyping. It's intended for multiplexing levels that are in the low end of the market, sort of in the 20 to 300 Flex range and then below 20 is where Eco is the right technology platform. So we think BeadXpress fills a good niche in the product lineup, and it's important for us to continue to focus on it.

Your next question comes from the line of Peter Lawson with Mizuho Securities.

Jay, just on the funding environment, what's the outlook for NIH spending from you and the benefit you saw in the quarter? And then secondarily, academic spending in Europe?

I guess, we characterize the overall spending when you integrate all the things that are happening around the world as stable. I think the best guess is in the U.S. is that the NIH budget will be flattish, could bump up a little bit, but we think an assumption of a flat budget is probably the safest one. Obviously, we'd love the budget to go up more, but the advantage of where we find ourselves is that we have the technology that is in the highest demand, and so we see money coming from other sources and being reallocated from other types of equipment and other projects, different agencies like the National Cancer Institute starting to dedicate more funding towards sequencing. So it's really the strength of the technology and the market we're in that makes the funding differences a little less relevant than it might otherwise be.

And then just on kind of the rare variant projects, how much of that do you think is actually migrating straight over to sequencing? Or are we too early for that?

A little to none. We really continue to believe that unless you're looking for Mendelian diseases, rare Mendelian diseases, then sequencing isn't going to be the answer for a long, long time. Arrays are way cheaper, way faster and frankly, more accurate than any sequencer that's out there. So we continue to believe that as long as the science supports the complexity that you have on the arrays, in other words, that there's confidence that the content of the array can deliver the right scientific result, that arrays will be the way that GWAS is done, not sequencing.

Your next question comes from the line of from Paul Knight with CLSA. Paul Knight - Credit Agricole Securities (USA) Inc.: Jay, what do you think the growth rate of the sequencing market will be for the five-year period ahead of us?

It's somewhere in the 20% to 35% range? Paul Knight - Credit Agricole Securities (USA) Inc.: And then the other is on the tax rate, you talked about Singapore. What's your goal on the tax rate?

Our goal is to get to sub-30, but we haven't given a specific time horizon. I do think it's achievable but we've got to move our manufacturing and we've got to continue to migrate our IP, particularly from the U.K. actually. A lot of our sequencing IP is based in the U.K. and we need to continue to migrate that to Singapore. So our goal is to get to sub-30, and I believe it's achievable, but we won't give a time horizon yet on this call. Paul Knight - Credit Agricole Securities (USA) Inc.: And then Jay, last, the two bottlenecks we get comments on would be on the back end, is data analysis a bottleneck you worry about? And on the front end, is sample prep a bottleneck?

Let me take the last part first. I think we anticipated the potential bottlenecks in sample prep because the throughput in these systems is going up to such a large degree that now you need to prep many, many samples, where before, one sample was plenty, because it took a long time to sequence one sample. And the progress we've made there is incredible in terms of reducing the time, increasing the indexing so you can multiplex samples together. And we're going to continue to move that front forward very fast in the next couple of years. So we don't think front-end sample prep is going to be a remaining issue. There's going to be a need to put more LIM Systems along the front end of the sequencers because now you have enough samples that you really do need automated ways of keeping track of them, where a couple years ago you didn't because there were so few samples going through. On the back end, I think we've done a great job of handling the analysis up through what we could classify as secondary analysis. And that can always get better. But what we're doing in the cloud environment now and being able to move data there and taking advantage of the elasticity of computing you get is going to be a huge advantage. We do think that the potential bottleneck that we all need to worry about next is figuring out how to utilize the data sets to extract the relevance of the underlying data to human biology. And that's a tougher problem because nobody knows exactly how to do that yet. The one thing that works in our favor is that in some ironic way, the larger the data set is, the easier it is to get statistical meaning from the data. And so while it's maybe difficult to figure out how to manage 10 genomes, if you add 1,000, the statistics pop out a lot easier if you have the ability to manage them in your computer system. And dealing with that problem is one that we don't think Illumina needs to be directly involved in necessarily, but it's certainly something we need to enable through partnerships, through collaborations, through investing in startups. And I think there's a great opportunity to advance the state-of-the-art there both in the commercial side but also through taking advantage of what's going on in academia.

With no further questions in the queue, I would now like to turn the call back over to Mr. Peter Fromen for closing remarks. You may proceed, sir.

Thanks, operator. As a reminder, a replay of this call will be available as a webcast in the Investor Section of our website, as well as to the dial-in instructions contained in today's earnings release. Thanks for joining us today. This concludes our call. We look forward to our next update in April following the close of the first quarter.

Thank you for your participation in today's conference. This concludes the presentation. You may now disconnect. Have a great day.