Monday, January 14, 2019

Illumina JPM Talk

Illumina CEO Francis deSouza delivered his J.P. Morgan Healthcare Conference talk (webcast audio, slides & Q&A audio) a week ago.  I can claim that some of my speculations came true -- just the most boring and obvious ones.  Overall, the presentation was the talk of a confident market leader.
On the sequencing front, deSouza spent most of the time talking about growth in applications and bullishly covering the many markets served with the instruments.  There were only two direct sequencing technology mentions in the talk.  For NovaSeq, the SP (for S-Prime or is it S' ?) flowcell -- which was announced last summer -- is the new focus.  This is now the smallest flowcell of the NovaSeq lineup, delivering 200-250Gbp of data in 2x150 mode.  The persistent Twitter chatter of a 2x250 mode on the NovaSeq were not discussed.  The other announcement is a 2x250 mode on the iSeq, to ship this quarter (but the specifications have not yet been added to the iSeq page).  He did mention that about 1/3 of MiSeq reagent sales are for the longer (2x250 and 2x300) kits.  deSouza stated that around 350 iSeq instruments have been placed, with about half of those with customers who did not previously own an Illumina sequencer.

deSouza did spend a bit of time discussing Illumina's strategy on the NovaSeq, which is perhaps one of the bits of sausage making scientists would prefer not to hear about.  Illumina has been orchestrating the release of flowcells in a pattern to encourage different sectors of the market to move off of HiSeq instruments and onto NovaSeq.  At one level this is simply good business: releasing all of the flowcells at once might lead to manufacturing issues.  But if you've been champing at the bit to get the SP, you might not want to know that its release timing might be more around marketing strategy than purely technical issues.

On overall genome cost, de Souza didn't really give a current cost but did say that Illumina is "marching" towards a $100 genome cost.  IN the Q&A he described Illumina as also waiting for the market to be ready for a $100 genome -- which is another comment likely to raise hackles in the research market by fueling the perception that Illumina is so confident in their market position that they can launch new technologies purely in terms of maximizing revenue rather than participating in a cutthroat marketplace.

An interesting trend that deSouza noted in the Q&A is the idea that national population sequencing efforts had initially started out as research and on specific diseases, but newer efforts are thinking in terms of starting in the clinic and aiming for population surveys from the outset.

Relatively little time was spent on the array business, which is seeing only single digit percentage growth.  Interestingly, deSouza credited the personal genomics market with much of the growth of the array business.  In the Q&A, he was asked if he can see microarrays ever becoming obsolete -- and while deSouza said he believes areas such as consumer genomics will move to sequencing but that there are markets in agriculture where microarrays will remain price competitive for a long time.

deSouza also touted Illumina's ready-to-go oncology panel, TSO500.  In the Q&A he addressed the difference between this and FoujndationOne, noting that Foundation Medicine performs additional services.  Illumina must walk a fine line here, trying to sell their own product without completely alienating a major customer.  deSouza also spoke of a second version of the non-invasive pre-natal testing product, which will detect more abnormalities in more chromosomes.

deSouza was very upbeat about population genomics efforts in multiple countries, such as England, France, Israel, Qatar and other countries.  This was one specific tie-in with the PacBio acquisition: deSouza spoke of such groups generating population-specific references on the PacBio Sequel and then scanning through entire populations on the NovaSeq.

Speaking of PacBio, as predicted by Mike Schatz there weren't a lot of details -- the deal has not closed.  Indeed, someone I spoke to last week familiar with antitrust regulation noted that the Federal Trade Commission is one agency affected by the ongoing partial U.S. government shutdown and that the Illumina-PacBio merger was one of many cases that appeared to have a pro forma regulatory clock reset just before the shutdown -- the FTC has a time limit to act but can restart that time limit by saying it needs more time to look. 

Otherwise, the message given on PacBio (with deal closure expected later this year -- my correspondent agreed there isn't any real antitrust case) will be to have the Illumina sales force driving the PacBio line, attempts to enable unified laboratory workflows and unified analysis.  It will be interesting to see how Illumina can unify the workflows, as handling large DNA generally requires much more careful handling than DNA going to short read platforms.  

When asked about nanopore, deSouza said they mostly compete in the long reads and emphasized that from their perspective nanopore lacks the accuracy required for clinical applications.

With regard to competition, the main talk was given almost as though competitors didn't exist.  BGI (as MGI) was brought up in the Q&A for their announcement of having 1000 instruments in the field, and deSouza first threw gentle shade by pointing out that Illumina has 13,000 installed instruments and then moved quickly to touting Illumina as the highest accuracy system around. 

5 comments:

  1. Thank you Keith, I think your previous article is more worth reading than Francis deSouza's talk LOL. By the way, I read some comments saying that some of illumina's core patents regarding to SBS / brige amplification, like US7972820 are about to expire soon, what do you think of that?

    ReplyDelete
  2. Just wondering whether there were many more paragraphs to go after the "BGI also ..." or was it a closing sentence ? ;-)

    thanks, and always appreciate reading the blog.

    ReplyDelete
  3. Ugh. Left behind a stub again -- will try to get something else out tonight to finish that thought.

    Not unexpected that some SBS patents are expiring; I should track that down

    ReplyDelete
  4. Does anyone have any info on ISEQ 100 real performance, from the lab? Are the users satisfied with it? Does it meet the expectations?

    ReplyDelete
  5. What have you heard about the initial performance of the five "Early Access" PacBio Sequel 2 (8M) systems? Does the Sequel 2 have the potential to be a game-changer?? The announcement from PacBio yesterday stated the Sequel 2 will be shipping in volume in "early Q2"

    ReplyDelete