ElysION vs. TraxION: Divergent Shots at Applied Market End-To-End Automation

London Calling was a particularly good opportunity to take stock of Oxford Nanopore's progress to a "fire-and-forget" sample-to-answer solution for "applied markets" such as food safety, public health and biotherapeutics quality control.   ElysION (formerly Project TurBOT) and TraxION represent very different approaches targeting different subsets of this broad market opportunity - and I heard from some interested parties that neither is quite what they want.  That doesn't mean they aren't right, but it does mean ONT may need to think of more approaches.

The broad concept is to have a a device that takes some sort of biological input, with minimal to no upstream processing, and performs all necessary steps so that nanopore sequencing data emerges from the instrument, with no human intervention after the run is set up.  ONT is envisioning these being placed in clinical labs, public health labs, biotherapeutic quality control labs, etc.  

ElysION

ElysION is a conventional benchtop liquid handler engineered by Tecan, with a moving gantry, a pair of heads on that gantry, and on-deck magnetic bead and thermocycling capability.  What distinguishes ElysION from any other liquid handler is ElysION has nanopore sequencing on the deck also, initially a trio of slightly modified MinIONs but later a P2 Solo will be an option.   On display was an instrument with a single MinION.  MinION flowcells were not designed with robots in mind, so ElysION has a custom jig which can be docked over the flowcell so that one of the heads can operate the access valve - and then that jig requires a specialized gripper in order to emplace and remove the jig.  Presumably a completely different jig will be required for the P2 version.  It's not the most efficient process from a minimizing-gantry-motion point-of-view (video below), but it works.

ElysION will require being set up like any other liquid handler - an operator must lay out the deck correctly with the right reagent holders and tip boxes, as well as load the correct flowcells in the MinION or P2.  And there will be the mirror-image cleanup at the end of the run - clear the deck of empty tip boxes and plates, remove flowcells, etc.

The first workflow targeted for ElysION is a bacterial workflow which takes lysed material as input and runs DNA extraction and then the rapid (transposon) barcoding prep.  Many others are planned.

ElysION certainly represents a sharp turn for ONT; they are painfully aware this is the largest and most expensive ($275K) instrument in their lineup, and not remotely portable.  But they see the applied markets as a major opportunity, particularly in areas such as RNA therapeutics where Direct RNA gives them a gigantic edge over any other sequencing platform - since no other sequencing company has commercialized a direct approach and its ability to detect RNA modifications.

The robotic lowcell loading aspect is one that many people I talked to loved - not only is learning to load flowcells not liked, but having to do it often is viewed as something that never gains any charm.  Plus ElysION will be capable of performing nuclease washes, enabling serial utilization of a flowcell for multiple samples - particularly valuable for Direct RNA since that protocol doesn't support barcoding.  Three MinIONs or one P2 should be suitable for many of the applied markets - but perhaps a bit underpowered for human genetics labs - though low-pass human genomes is on ONT's list of applications they are developing. ElysION workflows will support serial utilization of the flowcells - one could launch a protocol in the morning to load one MinION, then go again in the afternoon for another and launch a third the next morning. Or the instrument could load all three from one workflow.

One person pointed out to me an obvious sin against ideal automation - mismatch in utilization.  During most of a run, the fancy liquid handler will be sitting idle what the sequencer is churning away.  A clinical person commented they'd love the instrument - except the capacity is too low.  I can't help but wonder if ONT should offer an option to get two instruments, but with one configured to perform extraction and library prep, but the other to support only flowcell priming and loading, with most of the deck covered in ONT instruments - or perhaps just a PromethION 24.  That would be a better fit for clinical labs that want to run many samples in parallel - or perhaps in serial - but hate loading flowcells.  Of course, that's just me spitballing - don't ever confuse that with serious market research.

Conversely, one could imagine wanting many more small flowcells.  Even with the current footprint, the proposed "Grongle" of 8 clustered MinION MkII flowcells might be very interesting - particularly if a Direct RNA flowcell in the MinION MkII became available since Direct RNA doesn't support barcoding samples.  Perhaps Grongle in this conceptioni evolves to something closer to Plongle, the once talked about 96-well format sequencer.

ElysION will be offered as part of ONT's stable platform "Q-line".   ONT plans to apply for appropriate clinical device registrations for ElysION so that it can be used in regulated markets by 2026.  Nearer term, ONT is taking orders now for ElysION early access by year's end.

TraxION

TraxION is much more ONT's style - a portable unit with stylish lines with visionary ambitions.  TraxION is the latest attempt to find market traction with the VolTRAX electrowetting microfluidics platform - Oxford has pulled the previous iteration from their store.  TraxION is envisioned as the ultimate in low-touch automation - raw biological sample going in a port using no lab equipment (Clive Brown suggested a nasal swab or dropper) and then everything else happens with no user intervention; the TraxION cartridge would come pre-filled with all reagents.  An onboard MinION MkII (formerly SmidgION) would provide up to 10 gigabases of data.

But many devils lie in many details. How much will the cartridge cost and for that cost what single sample assays yielding 5-10 gigabases have a market?  Likely too little capacity for human genome assays, even using adaptive sampling to generate a virtual panel. Depending on costs, perhaps infectious disease, food safety and RNA therapeutic quality control markets would shift from ElysION to TraxION - ONT would probably lose little sleep over cannibalizing the market for the ugly ducking of their product line (and if you don't like my metaphor, keep in mind that ElysION isn't radically different in size than a full-grown swan) TraxION will be capable of magnetic separations and thermocycling, but what's the right number of reagent spots? 

The greatest uncertainty lies in getting TraxION functional.  The fluid link from the electrowetting operations surface to the sequencing flowcell is still being worked out.  Pre-filling the cartridge with reagents is very attractive to the user, but presents a myriad of challenges for the producer.  Clive Brown suggested that MinION MkII flowcells will be available for sampling by end-of-year, but that isn't a firm date nor is that enough to launch on.  So TraxION has no timetable yet.  

The MinION MkII will be very constraining on TraxION applications.  It's not hard to imagine that if successful, ONT would follow with a TraxION-like device that fed a PromethION flowcell.

TraxION-format MinION MkII flowcell, with a cartridge-loaded device at top of photo and a view of an unloaded one at the bottom.  The 4 spheres are magnets to support magnetic bead operations; above it is an optical sensor and below a thermal pad.  ONT has not settled on the linear arrangement of these active elements

Conclusion

So two very different takes on sample-to-answer sequencing.  ElysION marries proven liquid handling technology with ONT sequencers to provide a near-term, high capital solution.  TraxION is a grander and riskier vision of a device that requires nearly no training to use and low capital investment - and relies on a technology that to date has proven unreliable.  ElysION will also be easier to pivot to new applications, since it more loosely couples the reagents and sequencing technology.  Indeed, ElysION may be of interest to tinkerers, who wish to try automating their own protocols or perhaps design-of-experiments approaches to optimizing a given workflow.  Conversely, ElysION has many moving parts and will require a higher intensity of field service; TraxION when it launches will certainly have "sending a replacement" as the field service option.  

As noted above, there is also the question of whether the current concepts marry the correct sequencing capacity to the sample preparation capabilities - which gets into the always sticky issue of whether ONT's product lineup is too complex (my usual claim) or not complex enough to support markets (which I'm sure many of my suggestions can be taken as).  ElysION has two sequencer options proposed - I've thrown in several more.  Plus there is the question of flowcells - a key target market for ElysION is RNA therapeutic QC and ONT doesn't make Direct RNA flowcells for every flowcell type.  There is always a tension between simplicity and covering sufficiently enough of the fragmented nanopore market to drive significant revenue growth to the company; whether ONT can correctly thread this needle remains to be seen.

[2024-06-05 - a reader kindly pointed out a place where my text was ambiguous & also that I had a sentence fragment at the very end where I failed to complete a thought so several sentences were added to the final paragraph ]