Illumina Inc (ILMN) 2018 Q2 法說會逐字稿

Welcome to the Q2 2018 Illumina Earnings Conference Call.

My name is Daryl, and I will be your operator for today's call.

(Operator Instructions) Please note that this conference is being recorded.

I will now turn the call over to Jacquie Ross.

Jacquie, you may begin.

Thank you, Daryl.

Good afternoon, everyone, and welcome to our earnings call for the second quarter of fiscal year 2018.

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've 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 Francis deSouza, President and Chief Executive Officer; and Sam Samad, Chief Financial Officer.

Francis will provide an update on the state of our business, and Sam will review our financial results.

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 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 currently available information, 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 these documents that Illumina files with the Securities and Exchange Commission, including Illumina's most recent Forms 10-Q and 10-K.

With that, I will now turn the call over to Francis.

Thank you, Jacquie, and good afternoon, everyone.

The second quarter represented another strong performance for Illumina with record revenue of $830 million, up 25% from the second quarter of 2017.

Before we get into the detail, I'd like to thank the global Illumina team for the solid execution in the first half of 2018.

Genomic information is more valuable and actionable than ever before, and it is to the team's credit that we continue to move quickly, embrace new opportunities and find innovative ways to support our customers.

As a result, our business is growing as we enable a broad range of drivers, spanning applications, systems and geographies.

In the last 5 years, we estimate that over 225 petabytes of sequence data have been generated on Illumina platforms alone.

That volume of sequence data is roughly equivalent to 2.5 [million] (corrected by company after the call) human genomes sequenced at 30x coverage.

Despite this impressive repository of sequence data, to date, we have only ascribed significance to less than 1% of the variance catalog in the human genome.

So it's clear that the vast majority of opportunity remains ahead of us not only in terms of sequencing volumes but, even more importantly, in terms of advancing our understanding of the genome and our ability to positively impact human life.

I'll now update you on our business in the second quarter.

Reflecting stronger-than-expected demand, second quarter sequencing consumable revenue of $455 million grew 35% from the same quarter a year ago.

This included about $13 million of stocking associated with Chinese customers buying ahead of potential tariffs.

Excluding this, growth was 31%.

Second quarter sequencing consumable growth was broad-based, with customers in research, translational and clinical settings all contributing significantly.

We're seeing robust growth as our customers leverage genomics to drive discoveries, connect data to insight and routinely apply genomic tests in clinical and applied markets.

Whether it is to uncover new insight in Parkinson's or diabetes or to establish the utility of circulating tumor DNA for early cancer diagnosis, our customers are increasingly using sequencing as the primary tool to advance our understanding of the genome and improve patient outcomes.

Highlighting this breadth of drivers across our sequencing business, we saw consumable growth across all 3 sequencing system categories: high-throughput, which includes NovaSeq and HiSeq; mid-throughput, which is NextSeq; and low-throughput, which includes MiniSeq, MiSeq and iSeq.

NovaSeq consumables grew $40 million sequentially, with continued strong adoption of S4 and S2 and a healthier ramp in the S1 flows that launched in February.

As we expected, NovaSeq growth was partially offset by lower HiSeq consumables although the HiSeq 4000 consumables remained an area of strength.

Overall, sequencing consumable use is ramping faster than we expected.

In total, high-throughput consumables grew more than 35% from the same quarter a year ago.

It is now approaching $1 billion annual run rate.

And a sizable majority of HiSeq customers who have added at least one NovaSeq System have increased their total spend on consumables.

As a result, our full year expectation for NovaSeq-related revenue is exceeding the internal forecast we set at the start of the year.

It was also a strong quarter for NextSeq consumables, with growing demand across VeriSeq NIPT, oncology research, translational activity and clinical oncology testing.

Mid-throughput consumables grew more than 50% from the same quarter a year ago, demonstrating that the NextSeq continues to be an important sequencing workhorse, supporting a broad range of applications.

Pull-through per system exceeded the high end of our $100,000 to $150,000 guidance range.

We saw sequential and year-over-year growth in our lower-throughput sequencing consumables, which support our MiSeq and MiniSeq systems, with pull-through within their expected ranges.

Today, these systems represent more than 10% of our sequencing consumable business.

Finally, library prep grew 25% from the same quarter a year ago, driven by our core library prep portfolio, including traction with our new Nextera DNA Flex offering, which continues to receive positive customer feedback.

Our library prep portfolio is now approaching 15% of our sequencing consumable business.

Sequencing system revenue grew 10% sequentially to $123 million, up as we expected from a seasonally lighter first quarter.

We continue to be pleased with the broad appeal of NovaSeq.

Globally, more than 200 different labs have received the NovaSeq System, with demand driven by capacity expansion, HiSeq conversion and new to high-throughput customers.

The broad appeal and utility of the NovaSeq System is further reflected by the mix of flow cells customers are using, with most ordering a combination of 2 or more [available] flow cells.

Overall growth in NovaSeq consumables spanning S1, S2 and S4 has been faster than we anticipated, and the feedback and system performance continues to be positive.

We expect NovaSeq shipments in the 330 to 350 unit range for this year.

Moving to the rest of our sequencing system portfolio.

NextSeq shipments grew sequentially, with a good mix of capacity expansion and new-to-NextSeq customers.

Combined low-throughput system placements increased sequentially and year-over-year, with strong adoption by new-to-sequencing customers who represented over 50% of shipments from this group in the second quarter.

I'm pleased to share that we started shipping iSeq to customers in late Q2.

Early order activity has been robust with a good mix of new and existing customers in both academic and commercial settings.

As a standard Illumina practice, we will ramp production in the coming months.

Sequencing services and other revenue had another strong quarter, up 10% sequentially and 38% year-over-year to $106 million.

Roughly half of this total dollar amount is associated with traditional maintenance contracts.

The rest is primarily associated with our sequencing services labs, which support Genomics England in addition to some NIPT and RUGD customers.

Genomics England had a particularly strong quarter as it targets completion of its 100,000 Genome Project later this year.

Additionally, we also reported a contribution from our oncology partners, Bristol-Myers Squibb and Loxo Oncology.

Both programs completed their first full quarter of work and are progressing well in their codevelopment and commercialization objectives.

There has been a good collaboration between our team and that of our partners, demonstrating ongoing commitment to our shared vision of bringing transformative insights for targeted and immunotherapies to patients.

We are particularly encouraged by positive feedback from key opinion leaders on an early version of the TruSight Oncology 500 assay.

Moving to arrays.

Microarray revenue grew 25% year-over-year to $140 million, with growth not only in the direct-to-consumer, or DTC, but also in global research applications spanning genotyping and epigenetics.

Growth was strong in all regions, including China, where the launch of the Asian screening array helped drive a large number of iScan placements in Q2, which indicates future growth for arrays in the region as sites scale their operations.

We also secured our largest single order for the Infinium MethylationEPIC BeadChip as we drive larger epigenome-wide association studies in the translational research market.

While sequentially, microarray revenue was down associated with DTC functionality, we foresee robust demand for arrays as DTC broadens across geographies, applications and providers as precision medicine initiatives expand and research and applied uses for arrays increase.

Moving to regional results.

Americas revenue grew 19% versus the prior year period, driven by growth in sequencing consumables and microarrays.

EMEA continues to be an exciting market for Illumina and now represents about 1/4 of our total revenue.

EMEA revenue grew 36% from the same quarter a year ago, with a strong contribution associated with the Genomics England program that has now sequenced more than 70,000 whole genomes.

We also saw initial NovaSeq purchases from the Wellcome Sanger Institute as they prepare for their first 50,000 samples that will be sequenced as part of the U.K. biobank program.

Total Asia Pacific revenue grew 32% year-over-year with another strong quarter from Greater China, which delivered shipment growth of 42% from last year and 45% sequentially.

Even excluding the $13 million stocking order, China reported record shipments in Q2.

Illumina has a number of very strong sequencing service provider partners in China, and their growth continues to be driven by NIPT and research-based sequencing in a variety of areas, including oncology and reproductive health.

Outside of China, the rest of Asia Pacific and Japan, or APJ, was driven in large part by increasing volumes at our service provider customers and ramping ahead of precision medicine initiatives.

In addition to robust sequencing consumable growth, APJ also reported strong array growth, driven by a small but rapidly growing DTC customer offering and test service across Asia that is based on our screening array.

Before I hand the call over to Sam, I'd like to formally welcome the Edico team that moved to Illumina headquarters at the beginning of June and is already proving to be an integral part of the product development group.

Together, we are now focused on delivering the most complete offering of accelerated secondary analysis solutions for all our customers.

We're confident that our proprietary FPGA acceleration will improve time to answer on workflows across research, translational and clinical applications and look forward to updating you on the team's progress in the quarters ahead.

With that, I'll hand the call over to Sam for a review of our quarterly financials.

Sam?

Thanks, Francis.

As discussed, second quarter revenue grew 25% year-over-year to $830 million, driven largely by growth in sequencing consumables that once again exceeded forecast.

We also saw strong performance in both microarray and sequencing services and the contribution associated with our oncology collaborations.

Additionally, benefit from foreign exchange rates contributed 2% to the year-over-year growth.

Total instrument revenue in the second quarter was $127 million, down 7% year-over-year and up 8% sequentially.

Instrument revenue, therefore, represented 15% of total revenue in the quarter.

Consumable revenue was $540 million, up 7% sequentially and 34% year-over-year.

Services and other revenue was $157 million, basically flat sequentially and up 32% year-over-year.

In 2008, instruments represented 33% and consumables and other represented 67% of all revenue.

In the second quarter of 2018, consumables and services represented 84% of total revenue, highlighting how the mix of our business has evolved.

Moving to gross margin and operating expenses.

I will highlight non-GAAP results that includes stock-based compensation.

I encourage you to review the GAAP reconciliation of these non-GAAP measures, which can be found in today's release, and the supplementary data available on our website.

Please note that all subsequent references to net income and earnings per share refer to the results attributable to Illumina shareholders.

Non-GAAP gross margin of 70.3% was higher than expected and increased 50 basis points compared to the first quarter and 330 basis points compared to last year.

The increases were primarily due to higher mix of sequencing consumables, which represented 55% of revenue in the second quarter of 2018 and compares to 53% in the first quarter of 2018 and to 51% in the second quarter of 2017.

Non-GAAP operating expenses of $348 million were up $32 million from last quarter, largely reflecting higher R&D investments, headcount additions and increased variable compensation expense.

Keep in mind that first quarter operating expenses were lower than expected due to the timing of certain investments and headcount additions in addition to a favorable judgment on a patent suit.

Non-GAAP operating margin was therefore 28.4%, down from 29.5% last quarter.

Excluding Helix, operating margin was 30.6% compared to 31.9% last quarter.

The non-GAAP tax rate of 15.9% compares to 12.9% last quarter and to 25.1% in the second quarter of 2017.

Positively impacting the Q2 '18 tax rate was an additional benefit related to our Helix investment and the favorable 2017 foreign tax return adjustment.

Excluding the Helix benefit, non-GAAP tax rate would have been 16.9%.

For the second quarter of 2018, GAAP net income was $209 million or $1.41 per diluted share, and non-GAAP net income was $212 million or $1.43 per diluted share.

The impact of foreign exchange increased Q2 non-GAAP EPS by approximately $0.03 relative to last year.

Cash flow from operations was $295 million.

Improved revenue linearity led to a record DSO of 43 days compared to 47 days last quarter.

Capital expenditures in Q2 were $77 million, and Q2 free cash flow was $218 million.

We ended the quarter with approximately $2.5 billion in cash, cash equivalents and short-term investments.

Moving to guidance.

Total company revenue is now expected to grow at approximately 20% in 2018 to $3.3 billion.

This represents an increase of $124 million compared to the midpoint of our previous 15% to 16% guidance range.

Compared our previous guidance, the new growth expectation reflects higher sequencing consumables reported in the first half and an increase in the sequencing consumable forecast for the rest of 2018, notably associated with NovaSeq.

We continue to expect full year non-GAAP gross margin to be up modestly from 2017.

Non-GAAP operating expenses as a percentage of revenue are now expected to be slightly lower in 2018 versus the 43.5% reported in 2017.

This is primarily due to a higher revenue outlook and slower growth in headcount than anticipated.

Reflecting another more modest Helix benefit in the second quarter and the favorable tax return adjustment, we now expect our non-GAAP tax rate to be between 16% and 17% for the full year.

On a go-forward basis, you should expect a tax rate of approximately 19%.

We continue to monitor the tariff situation closely in China.

As noted earlier, 3 of our larger customers initiated some stocking purchases in the second quarter, and we expect to see additional stocking activity, albeit to a lesser extent, in the third quarter.

Beyond possible near-term variability in the timing of orders like we saw in the second quarter, we are confident that customers will continue to see the benefit of partnering with Illumina to deliver their own genomics-based services and products.

GAAP EPS is now expected to be between $5.10 and $5.20, and non-GAAP EPS is now expected to be between $5.35 and $5.45.

This increase primarily reflects the higher revenue projection for the full year, improved gross margin and a more favorable non-GAAP tax rate than previously projected.

For the full year, we are still expecting about $0.25 of Helix dilution.

This excludes the previously mentioned tax benefit that resulted from our latest investment round in Helix that positively impacted Q1 and, to a lesser extent, Q2.

We continue to expect share count for the full year to be approximately 148 million.

For the third quarter specifically, we expect revenue to be flat to slightly up from Q2.

Keep in mind the $13 million stocking order that effectively pulled revenue from later periods into the second quarter.

As a result, we expect a more modest sequential increase in sequencing consumables and we expect sequencing system revenue growth.

Offsetting this positive revenue impact in part, we expect arrays to be down meaningfully on a sequential basis.

This is primarily driven by seasonality in DTC, and in fact, we expect Q3 to be the lowest revenue quarter of the year.

And after the positive benefit we saw in Q1 and Q2 of this year, you should expect FX to have less of an impact than the back half.

We expect non-GAAP gross margin to be down modestly on a sequential basis to reflect the impact of mix, and we expect non-GAAP operating expenses to increase sequentially on a percentage of revenue basis given our expected hiring and investment plan.

With that, I'll hand the call back to Francis.

Thank you, Sam.

This was a very strong quarter for Illumina.

And looking ahead, the opportunity for genomics continues to grow.

With so much left to learn about the genome, research remains an important driver of our business.

In fact, 4 of our 10 largest sequencing consumable customers in the second quarter are primarily focused on research, either directly or as a CRO, and their combined demand grew faster than the overall sequencing consumable growth rate in the quarter.

The consumer genomics market continues to extend its reach, with some promising developments in China, Korea and Japan this quarter in addition to continued momentum in our DTC array customers and Helix.

In oncology, we saw Guardant Health receive the first -- the final Medicare coverage policy for an NGS-based liquid biopsy assay for certain non-small cell lung cancer patients.

We're seeing growing interest in tumor mutational burden, or TMB, which is emerging as one of the best NGS-based predictors of response to cancer immune therapies.

Germany is expected to be the first country to formalize reimbursement for TMB once the national study is completed in the fall.

And we've been seeing strong interest in our NextSeq system as a result.

We continue to see the NIPT opportunity growing with expansion into new geographies and broadening coverage.

NIPT was acknowledged as beneficial as a primary screening tool for trisomy 21 in Germany during the quarter, opening the discussion for potential reimbursement in 2019.

And in the U.S., we are watching to see whether ACOG will further revise its guideline regarding use of NIPT for average-risk pregnancies following the withdrawal of Committee Opinion 640 that was issued in 2015.

There's growing interest in population genomics initiatives, highlighting that governments all over the world are recognizing the value of genomic studies to enable precision medicine that could improve the quality and efficiency of health care systems.

In the second quarter alone, we saw All of Us began enrollment in the United States, a large U.K. biobank study awarded, and announcements from Genome Canada and the Australian government, among others.

While these programs typically take several years to ramp from initial interest to begin sequencing at scale, they represent exciting multiyear opportunities for us and our customers.

We'll now open the call to questions.

Operator?

(Operator Instructions) And our first question comes from Derik De Bruin from Bank of America Merrill Lynch.

Just a little bit clarity on this $13 million stocking order.

Is it across all the different sequencing categories, all the different sequencing instruments?

I'm just wondering if it was more heavily weighted to NovaSeq or if it was tied to -- or just spread across, just a little bit more color on what's going on to that market.

Sure.

The $13 million stocking order was primarily focused on NovaSeq, although it was a little bit of NextSeq in there, too.

And moving on to our next question, we've got Ross Muken from Evercore.

So maybe if we think about all the various areas of application or end market where we're seeing the sort of dramatic uptake on the NovaSeq side, where are you seeing the most sort of surprising level of interest in growth?

And if we think about kind of the mix today, clinical research and some of the more translational areas, what is that looking like on kind of an incremental basis as we think about where the growth upside from utilization is coming from?

So how is that mix changing?

Yes, sure.

There are a number of things that we expected to play out with NovaSeq that we are seeing, and there are definitely a couple of surprises.

One thing that we did expect to play out was that NovaSeq would fundamentally unlock some elasticity in the market.

And one of the things that we were looking for was validation that customers who purchased the NovaSeq, in addition to their other high-throughput systems, were -- ending up buying more consumables from us.

That would validate a theory that they would open up bigger cohorts, more access to samples, and that's certainly how it's playing out.

So that was sort of a core driver of NovaSeq, and then that's playing out well.

Obviously, we're seeing a lot of whole-genome work being done on NovaSeq.

As we expected, some of the big service provider customers of ours have been on the commercial side, certainly have been some of the earliest adopters and some of the fastest to ramp up the NovaSeq installation.

So we expected to see that, and that's playing out.

What maybe we didn't expect that we saw was quite the number of new to high-throughput customers that emerged in both the NovaSeq pipeline and then as NovaSeq customers.

And so that's been a pretty robust, consistent number quarter after quarter now.

And so that frankly was an upside surprise for us.

And as we talked about, the NovaSeq consumable pull-through has been also an upside surprise for us that customers are more quickly unlocking more samples and are more quickly ramping up their NovaSeq utilization.

And so all that sort of added up to us seeing that our NovaSeq revenues for the year across consumables and instrument is going to be greater than we expected for the year.

And our next question comes from Tycho Peterson from JPMorgan.

I want to ask about some of the benefits you guys had.

I mean, you quantified the tariff pull-forward, but I think we'll probably get questions about maybe some of the milestones around Bristol and Loxo and then GeL, which is going to tail off a bit in the back half of the year potentially.

So as we think about the guidance raise for the back half of the year, a couple things, can you talk about -- to the extent those are headwinds and what you're baking in for additional tariff impact, is there anything embedded for some of the newer products like Edico and the new flow cells?

And then how much of a headwind there are GeL and maybe some of these companion diagnostic deal milestones that won't necessarily repeat?

Yes.

So the -- there's definitely been some collaboration income -- or collaboration revenues, I should say, that we realized in Q2.

We expect that to continue in the back half of the year as well, Tycho, per our earlier guidance albeit probably to a smaller or lesser extent than what we saw in Q2.

We saw a good quarter from GeL.

We expect to see continued GeL revenues as well as they ramp up to achieve 100,000 genome milestones that they have.

So overall -- and in terms of the tariff question that you had, as we talked about in our prior prepared remarks, we saw $13 million stocking orders that we saw in Q2.

We do expect some timing shifts with regards to potential orders, placements of consumables ahead of potential tariffs in the back half of the year, but that's not

(technical difficulty)

material.

So there isn't really anything material to talk about in terms of major shifts because of tariffs.

So we saw that in Q2.

We expect to see maybe a little bit of it in Q3, but nothing really that material to impact sales.

And then with regards to -- I think your last point was around the NovaSeq consumables, the S1, S2, S4 flow cells that we have.

As we talked about in our prepared remarks as well, part of our guidance raise is driven by the fact that we expect increased consumables revenues from NovaSeq.

So it's reflected in our guidance range.

Yes.

That's right.

And if I were to add, if you think about GeL, clearly, GeL is working to wrap up the 100,000 Genome Project, and we're in the thick of that.

But GeL is also looking now to transition to the second stage of their journey, where they're actually building in genomic testing into the NHS and really moving into clinical standard of care for things like RUGD.

And so obviously, there'll be a transition.

But if we think long term, that could actually unlock a bigger revenue opportunity long term than the trajectory we're on now with them.

And our next question comes from Doug Schenkel from Cowen and Company.

I think Doug has dropped off.

We'll move on to Patrick Donnelly from Goldman Sachs.

Maybe just one on China.

I know you mentioned kind of the oncology and reproductive health markets kind of shaping up there.

Could you just also talk through -- I know consumer genomics is starting to pick up steam.

So maybe just give us the state of the market in China, which markets you're most excited about over the next 6 to 12 months there.

Yes.

So China continues to be, as you can imagine, a very exciting market for us.

It is already our second largest country market.

And as we look at the drivers, we feel very good about the prospect for continued growth in China, driven on the one hand by things that you talked about, which is our NIPT business still has had room in front of it there.

Obviously, oncology and oncology testing, we're in the very early stages of penetrating that market.

But there are a number of other levers that are emerging, and one that's definitely interesting is the direct-to-consumer opportunity.

And the launch of the global screening array has helped drive opportunities for DTC for us in China.

It's already led to a number of iScan placements in China.

And we have a number of companies that we're seeing now ramping up their DTC capabilities there, including WeGene, who is going to utilize our microarray technology to provide complete ancestry analysis service of the 56 major ethnic groups in China.

Another company in China, as an example, is working on an EWAS study in the second half of 2018, with the intention to launch a DTC-based study on -- offering on methylation.

And so we're definitely seeing the emergence of a DTC segment, although still at the very early stages, in China.

And our next question comes from Doug Schenkel from Cowen and Company.

Okay.

Can you guys hear me?

We can, Doug.

Okay, sorry about that, not sure what happened.

So as consumables revenue per NovaSeq has continued to ramp, HiSeq and HiSeq X utilization has dropped a bit but it's probably been a bit more durable than many of us expected.

Relatedly, HiSeq and HiSeq X retirements don't seem to have scaled a whole lot even as you place more and more NovaSeqs.

So with that in mind, over the balance of the year, what's your expectation for NovaSeqs impacting your HiSeq installed base of instruments and the use of these instruments that remain in the field?

And looking beyond this year, once more projects are completed, do you expect the pace of retirements to increase and the amount of spend for HiSeq and X to drop?

Sure, Doug.

So I'll start by validating a number of the points that you made.

One, we are pleased in the ramp-up that we're seeing in terms of the utilization of NovaSeq instruments that are out in the field, and we're pleased with what we're seeing in terms of pull-through of NovaSeq.

Although it's too early to call a definitive range, but we're pleased with what we're seeing so far.

We're also seeing though, in some cases, that there are customers that have validated workflows on HiSeq that will continue to run on HiSeq for a while before they transition onto NovaSeq.

And that's part of the design that we laid out in terms of having this multiyear transition into NovaSeq.

And so some of the dynamics that will play out over the coming quarters and years is we'll start to see some of those clinical customers validate their workflows on NovaSeq and transition over to NovaSeq.

We also see some of the larger HiSeq X fleets starts that cut over to NovaSeqs.

So they've begun that journey, but we're still in the early stages of people doing the complete cutover onto NovaSeq.

And so that's still in front of us, and that will drive the demand for NovaSeqs going forward.

I don't expect to see any sharp acceleration in that transition.

I expected to see it sort of more measured over the years, and that is how we've laid it out.

It's also how we staged the flow cells, so coming out with S2 at the beginning then S4 in the fall of last year and then S1 in Q1 of this year.

And each of those flow cells is intended to catalyze different segments of the HiSeq base, right.

So S4 is already targeting the high-throughput genome centers and looking to prompt, in the coming quarters, a cutover from X. S1 is obviously the lower end of that segment to move customers that might have been still on the 2500, for example.

And so that's how we've staged the roll-out of the flow cells, and that's how we expect it to play.

Yes.

I'll just add a couple of quick comments, Doug, as well just for completeness.

We've always said there's going to be a multiyear upgrade cycle, and that's how it's playing out.

And with regards to customers, they're transitioning at their own pace, and we're more than happy to see customers transition as they see the need to.

So that's how it's going to play out.

And our next question comes from Sung Ji Nam from BTIG.

Just not to belabor the point on tariffs, but was wondering in terms of the stocking activity, is that just out of abundance of caution?

And also, are you seeing the outside of China as well?

Yes.

The stocking activity that we saw in Q2 was purely related to China and for a handful of customers in China and as Francis mentioned earlier, mostly on NovaSeq, some orders of NextSeq as well.

So at this point, we're seeing it in China only.

And truth of the matter is, we're still monitoring developments on the tariffs point and what's going to happen.

So we're still looking forward to see what's -- how this is going to play out.

As I said, we don't expect major shifts in terms of ordering and placement activity.

We see this to be fairly immaterial in terms of our overall sales.

And the last thing I'll say on this is sequencing technology is a big investment and one that is not easily redirected.

Our customers take this seriously.

And so as we look forward and as we stay and stay close to this whole tariff situation, obviously, we're going to continue to watch it closely, and we're going to also do what's best for our customers as we look forward.

Yes.

So just to add to what Sam said, look, ultimately, the demand in China is driven by the customer demand and the strength of our technology.

And we continue to expect both to be strong and sort of drive the business for us there.

And our next question comes from Dan Arias from Citigroup.

Looking to see, Francis, if maybe you could provide the revenue or the system growth number for oncology during the quarter as we sort of track momentum there.

And then maybe more broadly but along those lines, I know it's hard for you guys to pinpoint application sometimes.

But as we're sort of contemplating this inflection that seems like it's taking place, I'm wondering if you could just sort of help us sort of what percentage of revenues, maybe roughly speaking, are now coming from clinical work.

How much falls into that bucket at this point?

Yes.

So we don't actually break out the revenue down to the level of systems placed in the oncology market.

But what I can say at a very top level is that the percentage of our overall revenues that come from the clinical market last year was about 45%, and that continues to be strong.

And so you've seen that grow now over the last few years as both oncology and NIPT have driven the growth of both consumables and system placements into those clinical markets.

Some of the things that are playing out in the clinical market, we continue to see strong translational demand for our products.

And then more recently, as we announced the companion diagnostic deals with Bristol-Myers Squibb and Loxo, that's starting to generate revenue and we expect that to continue.

And then one exciting development for the future is the emergence of tumor mutational burden as a biomarker predictive of the effectiveness of immunotherapies.

And that could be an exciting driver to add fuel to the oncology market for us.

And our next question comes from Bill Quirk from Piper Jaffray.

First off, Francis, I really appreciate the color around both Genome England as well as Sanger's efforts around population sequencing.

Can you talk to some of the other country programs and help us think about the pacing of some of these programs and the associated spend?

And I have a follow-up.

Yes, sure.

There is no doubt that these population sequencing initiatives are really exciting, both for the market and for Illumina.

These programs, they deepen what we know about the genome.

They have the potential to enable precision medicine at a national scale.

And perhaps equally importantly, these initiatives, they educate the population about genomics and the value of genomic information in our lives.

So hugely exciting initiatives and it's fantastic to see the momentum these are gaining around the world.

Now having said that, these are large initiatives and they can take years to scale.

You brought up GeL.

GeL started in late 2014, and it's really only now that it's at scale as it's looking to complete the 100,000 genomes.

So hugely exciting for a number of reasons and we're doing a lot of work, as you can imagine, to support the initiatives, to support the customer tenders, to support the pilots.

But it's not something that we're counting on as a dramatic revenue driver in 2019.

And our next question comes from Jack Meehan from Barclays.

I was hoping you could elaborate on some of the early traction you're seeing with the iSeq launch and just elaborate on what your expectations are in terms of placements, what are some of the customers you're seeing look to adopt the technology.

Then finally, if your thinking has involved at all around how this could be used as a clinical tool.

Yes, sure.

I'm particularly excited about iSeq and the role it plays strategically in our portfolio.

When we envisioned iSeq and then when we started talking to the market about iSeq, one of the things we said is that there are tens of thousands of labs, smaller labs, that today don't do any sequencing.

And so we see the opportunity there to use iSeq to bring sequencing to those labs.

And we talked about the fact that we expect iSeq customers to fall in 3 buckets.

We said, look, some of the early iSeq customers will be existing Illumina customers.

And the use cases they'll be drawn to is to use iSeq as a QC tool before they run their libraries on the bigger machines.

If you think about a run on a NovaSeq, for example, that run is tens of thousands of dollars.

And so customers are excited about having iSeq to QC the library before they run it on NovaSeq.

And sure enough, if we look at the early orders for iSeq and we look at the pipeline, there are definitely customers that fall into that bucket.

They're excited about having it as a companion to their higher-throughput instruments.

The second bucket though, we said, is there may be labs that are familiar with sequencing but don't have a sequencer because they're outsourcing their sequencing.

And they're doing that for a couple of reasons.

One, they may not have access to the capital necessary to go buy even a MiSeq but certainly not a NextSeq or a NovaSeq, or they just don't have the regular flow of samples.

And so iSeq is a really exciting tool for those customers.

And again, if you look at the orders we've got and the pipeline we're building, there's definitely a set of customers that are unfamiliar with sequencing but don't have a sequencer today.

But then third, we said -- and probably most exciting are going to be the customers that don't do any sequencing today that maybe used a qPCR machine or some other tools that iSeq can help address.

And again, as we look at who's buying iSeq, we're definitely seeing a number of customers that are new to Illumina and, in some cases, new to sequencing in total.

We started thinking about the applications that we're hearing from our customers in addition to the companion QC tool.

We're hearing customers that are looking to use the iSeq for things like pathogen detection, for microbial whole-genome sequencing and for targeted RNA.

Those are probably the use cases we hear the most at this point.

And our next question comes from Puneet Souda from Leerink Partners.

I have a broader, more longer-term question on NovaSeq consumables.

You have obviously seen a great pick-up here in consumables and the guide is reflecting that.

So I was wondering, when you look at it on a multiyear replacement cycle that you have pointed out for NovaSeq, historically, for HiSeq, we had seen utilization ramp-up and reach a certain level of peak.

But in case of NovaSeq, should we assume the utilization and/or the pull-through to continue to ramp up longer term, assuming that potential flow cells will come to the table faster than potential new instrument launches?

Yes.

If I think about NovaSeq -- and you're right, the trend we saw before was the consumables started to build and then hit a peak.

And then once we hit a steady state, we were able to talk to the market about a range that you could use to model pull-throughs.

We are still at the stage where the NovaSeq pull-through is continuing to go up.

In fact, actually, in the last quarter, the pull-through on NovaSeqs hit a new high.

And so we're still at the stage where it's building and frankly more than maybe we'd initially expected even.

At this stage though, it's still volatile.

And so we're not yet at the stage we're able to give you a range that you can use in the models.

And the scenario you brought out -- brought up is sort of a realistic scenario, which is customers are using multiple flow cells and they're mixing and matching the flow cells that they're using for the applications they want to run.

And NovaSeq is a more versatile tool than something like a HiSeq X, which is clearly a whole-genome machine.

And so I think that creates more utilization opportunities for customers than a HiSeq X.

And our next question comes from Amanda Murphy from William Blair.

So I just had a follow-up on the population sequencing commentary that you've made.

So it seems like some of these projects -- obviously, there's been an increase in countries and other entities coming online.

But it seems like there is some conversation around the right methodology, I guess, between genotyping and sequencing and, I guess specifically thinking about all of us, where they discussed price point and what they might view over the long term.

So I guess my question is, as a -- do you have -- obviously, you're having conversations with the customer base in that vein.

Do you have that kind of a current instrumentation platform that you need to sort of get these guys over the hump in terms of doing broader whole-genome sequencing, meaning the S4 and et cetera?

Or is there an evolution that we should expect before some of them turn to big whole-genome-type projects?

Sure, Amanda.

So I'll start by saying there are now dozens of POPSEQ opportunities that we are working with around the world, and they're all a little bit different.

Some draw some common trends across some pretty different opportunities in terms of size and scale.

I'll start by saying that one common thread across the majority of them now is that they are connected to a health care system directly, even more so than we saw when GeL pioneered population sequencing in 2014.

And so the vast majority of those opportunities now are either directly connected to the health care system or actually driven by the health care system.

The other trend we're seeing is that the majority of them are around whole genomes.

All of us started as sort of thinking about a 2-stepper, where they would start with genotyping and then in a future stage, they would go to sequencing.

And what's happened now is they've accelerated their sequencing.

And so now they are talking about doing both simultaneously.

So even all of us, where I was initially genotyping and then going to sequencing, it's really brought sequencing forward.

And I think there, the -- overall, the work that's happened at GeL has really been helpful for the rest of the world in terms of seeing how it could be done and seeing the utility of doing whole-genome sequencing.

In terms of our platform's readiness, S4 is absolutely appropriate for the large-scale population sequencing efforts that we're talking about, and I think it can carry us for a while.

Obviously, in the longer term, as we think about the scale of how these -- how big these initiatives could be as they get embedded across multiple applications in the health care system, you'll need a bigger boat.

But for now, S4 is absolutely the right answer.

And our next question comes from Dan Leonard from Deutsche Bank.

So I was hoping you could elaborate on your DTC assumptions and the visibility you have there.

And specifically, I was surprised on the Q3 comments that you would expect sequential weakening given the amount of promotional activity on Prime Day, the adoption of WeGene and the overall earlier stage of penetration in Asia and some of the other demand drivers.

Yes.

Sure, Dan.

I'll speak to it.

We haven't given any expectations for the full year in terms of DTC or microarrays.

So we haven't updated our assumptions there.

But the seasonality that we talked about is really not uncommon.

We see that seasonality every year, where we have a strong Q1 and then, to some extent, a lesser Q2.

Q3 is typically the -- I would say the weakest quarter of the year for DTC and hence, why we have indicated that for microarrays in total, it's going to be a weaker quarter.

And then we see potential improvements in Q4.

That's the usual seasonality that we see with microarrays and DTC.

And this year, we don't expect anything different.

And our next question comes from Steve Beuchaw from Morgan Stanley.

I'll come back just to the Nova consumables question but from a couple of other angles.

One is when we think about Nova consumables and the efficiency that it provides to labs, they give people more for their money, right.

That's pretty straightforward.

And so I wonder, is it right to think about that sample volume is really picking up at a pace that's materially more than the 30-plus percent rates that we're seeing in the revenues?

And is that something that you would see in BaseSpace from the operations out there in San Diego?

Do you see revenues -- I'm sorry, samples actually growing quite a bit faster than revenues?

And then I wonder if you could circle back on the same topic to sample access.

When I think back to 2014, 2015 and we had the interest in X emerging, there was, at some point, difficulty getting to samples -- enough samples to keep them running at capacity.

But this time around, it seems like the -- these -- the access to samples is going very, very fast.

Is the key there that it's not just whole genomes this time?

Or is there some other key to the explosion of access to samples?

Yes.

What we are hearing from our customers and we're seeing in the pull-through, what we're seeing in the instrumentation data is that customers are running their NovaSeqs hard, and that's showing up in the utilization numbers that we talked about.

That's showing up in the pull-through numbers and the highs that we're hitting in terms of pull-through on NovaSeqs.

And what is playing out falls into a number of buckets.

One, you're absolutely right.

We are seeing our larger cohorts being run and where they're being run.

Second driver we're seeing is there is an appetite for doing broader sequencing.

Things like going for exomes instead of panels with the biobank or going for genomes instead of exomes is definitely playing out.

And then some of the other drivers of people to purchase NovaSeq include applications in oncology, like looking for rare events associated with cell-free tumor DNA in the blood.

And for there -- for those applications, what NovaSeq allows you to do is it allows you to go very deep.

And so those customers, for example, find NovaSeq very attractive, and they run them harder (inaudible) they could most other machines.

And it's very useful for them in terms of being able to understand the biology.

And so what NovaSeq is doing is it's fundamentally enabling new paradigms for experiment design there in a way that couldn't be done without the power of NovaSeq and without access to the economics of NovaSeq.

And so all of those are playing out.

And what's interesting about the NovaSeq dynamic, if I compare it with the X dynamic, is -- and I don't think you're on a point there, which is with the X, it was really targeting whole genomes.

And so that was the application.

And if you were interested in that application or if you wanted to move from exomes to genomes, the X was a terrific machine for you.

But with NovaSeq, we enable a much broader set of scenarios.

And so if you want to do exomes or you're doing panels, you wouldn't have been able to do that in the X. You can do that here.

If you want to do very deep targeted sequencing, again, that's something that's uniquely for NovaSeq, not for the X. And so it's that breadth of applications, as you pointed out, that means that our customers can access a much broader set of samples, a much broader set of applications, a much broader set of experiment designs.

And that helps them with getting access to samples.

And our next question comes from Mark Massaro from Canaccord.

Francis, can you speak to the inflection in new research studies, for instance, in single-cell sequencing?

And then related to your comments about exome sequencing, I know Geisinger indicated plans to offer whole-exome sequencing to their 3 million members.

Do you expect that to play out later this year?

And do you expect other systems to potentially follow suit?

Yes, absolutely.

So as I said touched on the last answer, NovaSeq is enabling the fundamentally different types of experiments, much, much larger experiments.

I know for example, if you talk to the people that are working on the single-cell atlas, for example, they will talk about the fact that if you want to catalog all the different types of cells in the human body -- and there are, what, 37 trillion-ish cells in the human body, you need to design experiments that are in the tens of millions of cell.

And that's got to be the design.

And they've been talking to us for a while, saying that they needed a machine that ended up being something like the NovaSeq, they said.

And you need a machine that's as powerful as a NovaSeq with the economics of the NovaSeq to even design and run that experiment.

And so we're enabling -- whether it's much bigger cohorts like the single-cell atlas we'll have access to or much deeper sequencing, these are experiments that can now be designed and run in a way that couldn't be done without the NovaSeq.

So absolutely, NovaSeq is driving new experiment with design paradigms that -- than unlock elasticity associated in the market.

Geisinger.

And then in terms of what we're seeing from the health systems, I think Geisinger is definitely a pioneer in terms of going first but I fully expect over time, your genomic record to be a fundamental part of your overall health record.

And so it will take time to go from where we are to get there, but Geisinger is not going to be the only one, I believe, that starts to do that in the coming quarters.

In terms of the pace at which that plays out -- and frankly, I think we won't stop at the exome.

I think it will be the whole genome that's part of your health record, but that will take time to play out and that will take pioneers like Geisinger and the other ones that are thinking about doing that to lead the way.

And our next question comes from Catherine Schulte from Baird.

We've seen varying levels of year-end budget flushes over the last few years.

So Sam, can you just walk us through what kind of assumptions you have in guidance at this point for fourth quarter research spending patterns?

And then Francis, any other commentary you're hearing from research customers around the current funding environment?

Yes.

So we're basically expecting what we usually expect around research spending patterns, nothing that's really different from the prior trends that we're seeing.

So in other words, without getting into too many deals around guidance and what's built into that, nothing really unusual in terms of budget reduction or anything that impacts research portion of our revenues.

Research actually has been showing strong growth for us.

The research component of our revenues, clinical, has been showing growth for us as well, and we expect that to continue, so nothing built in around any unusual budget crunch at the end of the year.

And in terms of the overall research funding environments, if we look around the globe, the research funding environment has never been frankly this good.

If you look at what's happened with the NIH budgets here in the U.S., if you look at -- there was a question about how Brexit would play out.

But if you look at the funding playing out both in the U.K. and Europe and if you look at what's happening in terms of overall research funding in China, you add it all up to a robust research funding environment.

And our next question comes from Derik De Bruin from Bank of America Merrill Lynch.

Just -- you're now seeing, what, $1.2 billion in net cash.

I'm just curious as to -- can you talk about -- sort of like thinking about capital deployment, I just think -- and obviously, you bought Edico.

But I'm just wondering, just put in a broader context, I mean, your cash flow generation is remarkable at this point.

Yes.

Thanks, Derik.

I'll speak briefly to it.

So anything that lowers the barrier for the adoption of genomics is one of the key, I would say, foundational elements of our capital deployment strategy.

So if it's technology that we can buy rather than build that helps our customers with the end-to-end workflow around sequencing, that lowers the barriers for the adoption of genomics, that's really going to be critical to us and very fundamental in terms of how we deploy capital towards M&A.

So that's, I would say, one key component of capital deployment.

We're going to continue to look for potential technologies that make sense to rather than build for many reasons.

Edico was a great example of that, as you saw.

And then the other drivers of capital deployment or the other, I would say, priorities offsetting equity dilution, as we've talked about time and time again, we continue to look at potential share repurchases opportunistically to offset equity dilution.

And then keep in mind, we have 2 tranches of convertible debt that come due in 2019 and 2021.

And so that should be in the backdrop of your thoughts around also as we think about available cash.

And our final question comes from Bill Quirk from Piper Jaffray.

Given the landmark CMS decision around oncology -- NGS oncology reimbursement earlier this year, we've noticed an increase in activity at the clinical level, specifically in hospitals.

And Francis, can you speak to the interest within systems?

And I'm thinking specifically around the community-level hospitals rather than the academic medical centers.

Sure.

We definitely agree that we are seeing sort of continued building of the interest in oncology testing.

Definitely, there's building interest from the translational perspective.

Now a lot of that tends to happen at the academic cancer centers, the larger cancer centers.

But we are starting to see interest from some of the smaller hospitals, too.

I think at this stage, probably the way that might play out is more of it will be driven by outsourced sequencing.

And so maybe some of that demand will be showing up first in some of the outsourced labs.

But over time -- and certainly, as you look at our NextSeq on -- down in the portfolio, we think those would be great instruments to put into community hospitals for -- to run oncology panels, for example.

And so we think that, that interest continues to build.

It's still in the vast majority of that market.

However, it's still in front of us, and we expect to -- for that to show up in the mid- and low-throughput part of our portfolio.

And that concludes our question-and-answer session.

I'd like to turn the call back to Jacquie Ross for closing comments.

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 instructions contained in 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 third fiscal quarter.

And thank you, ladies and gentlemen.

This concludes today's conference.

Thank you for participating.

You may now disconnect.