Oxford Nanopore @ ASHG

[ymc]

I just noticed Oxford Nanopore thinks solid state nanopore is the future. They added Harvard's solid state nanopore team in 2011. If this new team develops fast enough, the current gen of protein nanopore might just be a stopgap product?

https://www.nanoporetech.com/technology/introduction-to-nanopore-sensing/solid-state-nanopores

[Jeremy]

Unfortunately my email was:

Thank you very much for taking the time to apply to join the MinION Access programme (MAP). We received a large number of applications from a broad range of applicants and unfortunately cannot invite every applicant at this time. Unfortunately we cannot offer you a place in the first wave of MAP however we intend to make more places available in the future and hope that you are happy to remain on our list of interested participants.
Once again we thank you for your interest. If you would prefer not to be considered for MAP in the future please email us on [email protected].

I hope those of us that did get in keep everyone posted in this thread.

[gringer]

There's a fair bit of information on the Nanopore website now:

https://www.nanoporetech.com//technology/the-minion-device-a-miniaturised-sensing-system/a-guide-to-map

Here's my rough summary:

• Restricted open-access philosophy for results and software development via login at nanoporeonline.com
• 6 week MAP cycles, 2 weeks for configuration, 4 for your own experiments. Burn-in cell in week 1, two experimental flow cells delivered week 3, two more week 5
• Can duck out at any time with a deposit refund
• Can continue on in the program for the next cycle, or stop and give someone else a go
• Cost is $1,000 deposit (covers all MAP cycles) + $250 delivery (per cycle for multiple deliveries & returns) + taxes where relevant
• ONP receives diagnostic statistics via the Grouper instrument software, but results are yours to keep
• ONP requests that people remain tight-lipped until the end of the burn-in period, so people will be quiet about this for at least the next couple of weeks
• Computer requirements are not mentioned on this page, although it is stated that Macs are not yet fully supported

[ZFHans]

The invitation is specific to the point of: "we look forward to working with you in the coming weeks."
More details will follow on the portal website that they have set up.

[GenoMax]

@NextGenSeq and @ZFHans: Does the "invitation" say when you are likely to actually get the kits?

[NextGenSeq]

We got an invitation too.

[ZFHans]

At least we can now start to judge for ourselves.
We got an invitation to join MAP and I suspect I'm not the only one.
I'm hoping we can soon produce some real data and make comparisons based on actual facts and not speculation.

[ymc]

Some of the VCs may not fully comprehend the full barriers to entry over and above the sequencing itself (sample prep, informatics pipelines, understood error profiles, etc) but they are no dummies. They had a new funding round very recently, so unless they completely misrepresented themselves to their investors there must be something there for real. I can't believe that anyone would give them any new money at this stage if they cannot produce some type of sequencing-related data. For example, PacBio did suck in $600M of VC money that will probably never be fully recouped, but they were not all smoke and mirrors and have mounted a niche comeback as of late.

I am sure the VCs saw nice thing when they put in money last round. However, they are likely to be caught in surprise for this mini-breakthough of HiSeq X that churns out 10x the throughput of HiSeq 2500. It seems now more and more likely nanopore can only be niche player that might only offer the VCs a break even return at best.

[BBoy]

I suppose their VC backers are wetting their pants now.

Some of the VCs may not fully comprehend the full barriers to entry over and above the sequencing itself (sample prep, informatics pipelines, understood error profiles, etc) but they are no dummies. They had a new funding round very recently, so unless they completely misrepresented themselves to their investors there must be something there for real. I can't believe that anyone would give them any new money at this stage if they cannot produce some type of sequencing-related data. For example, PacBio did suck in $600M of VC money that will probably never be fully recouped, but they were not all smoke and mirrors and have mounted a niche comeback as of late.

[ymc]

There was a per-Gb cost estimate for the minion that somebody posted, I just can't find it right now. It was ~10x that of a Hiseq and ~2-3x that of Pacbio, sample prep and reagents not included. That was based on the stated flowcell price of $1k and a run time of about 6 hrs, which is the only hard information available about ONT to the general public. There has been zero talk about the gridion even from ONT, so I don't know how their previous claims can be taken as anything more than a marketing spiel.

Let's give them time to actually show that they can sequence and then speculate how they will take over the world. When expectations meet reality, it is usually the latter that retains its reputation. If their early access program is for real some actual information should start coming out this year... unless the fine print at the bottom of the early access application reads "please allow 3-5 years for processing"

Yeah. But even that marketing spiel years ago doesn't sound that good any more when HiSeq X spec is announced. Assuming they can come up with it in two years in the most optimistic scenario, Illumina will go to another gen. It seems more and more likely after waiting for so many years, they might still not be able to dethrone Illumina except for applications requiring long reads and minion for outdoor use.

I suppose their VC backers are wetting their pants now.

[BBoy]

Google tells me even for GridION 8000, their reagent cost is $20-30/Gb.

But HiSeqX is $800/90 = $8.89/Gb

Certainly their $30K for the box claim can make up a lot of ground

There was a per-Gb cost estimate for the minion that somebody posted, I just can't find it right now. It was ~10x that of a Hiseq and ~2-3x that of Pacbio, sample prep and reagents not included. That was based on the stated flowcell price of $1k and a run time of about 6 hrs, which is the only hard information available about ONT to the general public. There has been zero talk about the gridion even from ONT, so I don't know how their previous claims can be taken as anything more than a marketing spiel.

Let's give them time to actually show that they can sequence and then speculate how they will take over the world. When expectations meet reality, it is usually the latter that retains its reputation. If their early access program is for real some actual information should start coming out this year... unless the fine print at the bottom of the early access application reads "please allow 3-5 years for processing"

[ymc]

It can certainly compete on equipment cost and minimum run cost. ONP has previously claimed that GridION will have sub-$1000 genome sequencing (and not just for human), so I'm guessing yes.

Google tells me even for GridION 8000, their reagent cost is $20-30/Gb.

But HiSeqX is $800/90 = $8.89/Gb

Certainly their $30K for the box claim can make up a lot of ground

[gringer]

It can certainly compete on equipment cost and minimum run cost. ONP has previously claimed that GridION will have sub-$1000 genome sequencing (and not just for human), so I'm guessing yes.

[ymc]

Now HiSeqX is announced. Can Girdion compete based on its paper spec announced so far?

[GW_OK]

Is anyone in the early access program?

Karen James @kejames
@SFriedScientist I learned a little more about @nanopore’s MinION today from someone at @jacksonlab who is going to get early access. Squee.

I'm guessing the fix is in for a number of early access people, most likely those 40-ish who got to actually see one running at ASHG, plus a few other of the twitterati who seem to know things before most.