comstock

Yes, but I would say quality is most important in almost all cases. Well, quality being defined as <1% error rate, which isn’t such a high bar.

The most compelling near term applications (NITP etc) use fragmented DNA, and long reads will have no benefit here.

So, yes. Long reads are useful, but you need to have at least reasonable performance in other respects. The same thing has been seen with PacBio, who have not played well in the market, despite having a read length advantage.

Do you have a citation for this? Because this is not my understanding.

Most clinical applications don’t need long reads. Pathogen identification from short reads is easy. Blood tests for cancer, and NIPT (which will likely be the first big applications) both use fragmented DNA in the blood, so long reads are not useful. Depth (lots of sequencing) and quality are far more important.

They’re going to need more money.

I think they’d need to sell >10,000 units a month. That is if they are making any kind of profit. Unfortunately I don’t think there’s a market for anything like that number of flow cells.

I think they like to say they have a patent monopoly, but I don’t believe they do. Genia (which also has issues) are developing a protein nanopore platform.

Illumina (see recent IP battle with Oxford Nanopore) also have key IP in this space (for mspA).

I would guess they currently make a loss on every device shipped. Most current sales are about generating press coverage/traction, not revenue.

The accounts of all UK companies are filed publicly. They had about 1M GBP of sales in 2016 from memory. I would guess with overheads, each unit is currently sold at a loss.

Illumina doesn’t beat nanopore because of the “scale” it beats it because the technology produces fundamentally higher accuracy data. The error rate on Nanopore reads is >10%. Even on first generation Illumina machines the error rate was 1%, and is now significantly lower.

Oxford Nanopore have been working on this for 15 years, this is the best they can do. It might be possible to create a Nanopore (maybe solid state, not protein) system with a lower error rate, but I think they have little hope of doing it.

That might be the case if the throughput wasn’t so low and the error rate wasn’t so high.

I’d agree with you, that long reads would be useful if the error rate wasn’t so shockingly bad.

There is, likely value in long reads, but what non-niche research applications are there for highly error’d reads that justify a valuation of several billion dollars?

NextSeq is expensive, but there are cheaper options (from 30k) and you can just send your samples to a sequencing service. I’ve seen costs for a whole human genome at high coverage of between 1 and 3000USD.

That’s great, but for a company that has raised on a valuation of >2BUSD how are they going to compete with Illumina.

The read lengths might be long, but the error rate is a couple of orders of magnitude higher. I can’t see a market large enough for highly error’d log reads to support their valuation.

It’s also comparatively expensive compared to other platforms (you can get a full human genome sequenced at high coverage for between 1000 and 3000 USD).

The error rate is stupidly high (somewhere between 10 and 20%) compared to Illumina or Ion Torrent who give error rates far less than 1%.

It can give very long reads, which are useful in some niche applications. But it’s been massively over-hyped (and over capitalized).

The neat thing is that it’s very small. But that isn’t really compelling given the very low accuracy.

Work for a year then take two years off? Or retire earlier? Again, I didn’t say that people wouldn’t work at all. I said that the barrier to obtaining a personal autonomy is too high in a society where providing the essentials of life has been largely automated.

We keep people working by making it had to survive otherwise (high rents that don’t represent the maintainence/construction costs etc).

Can be useful when you’re looking after kids, you’re carrying a child/stuck away from your phone.

I’d guess voice assistance could be useful in other situations too. But for me the error rate still seems too high, and there are limited applications.

I’m not really talking about providing a basic income.

What I’m saying is the barrier to having “personal autonomy” in a society where providing the essentials of life has to a high degree already been automated, is too high.

The last sentence really stuck home to me (edited for clarity):

While these Red Ocean ideas may not result in creating multibillion dollar companies they do offer opportunity that can lead to riches and personal autonomy.

It’s sad that we live in a world where the goal of personal autonomy is so lofty. We have an over abundance of resources in the West. And yet having the freedom to work on useful things under our own direction is rare. It’s a shame that having “autonomy” is such a difficult goal to obtain.

Only if you have multiple devices... it would be nice if they supported this optionally (it would be nice if they supported any kind of end-to-end backup optionally).

It’s an AI using deep learning and therefore magic requiring no explanation.

I think they want to support the scenario of lost device, lost password, but still allow restoring data to a new device.

Yes, they obviously have the infrastructure to do it. However, they don’t even optionally.

Part of the reason is that people sometimes forget their passwords and that would lock them out of their backups. So they want to allow email/other methods of resetting the password and giving access to data.

But it would be nice to have it as an option. It’s worrying though that even technical people seem to believe it is end-to-end encrypted. When it very obviously isn’t.

I probably could have been more tactful. However the substance of my comment is correct, and while I believe Apple do great things in security and promoting privacy rights, they don’t provide end-to-end backups for most data via iCloud.

They don’t claim to, but the same misinformation gets spread every time this topic comes up on HN. If you want end-to-end encrypted cloud based backups of photos and other data on iOS you presently need to use a third party app.

Would you care to link to the section of the security white paper that says that? There’s a small subset of data that is encrypted end-to-end. Your photos, notes, most and most other information is not include in that.

It’s pretty obvious really, they need to know the key for encrypted at rest data in order to be able to reset your password if you desire. They absolutely do don’t currently offset end-to-end encryption on the majority of data in iCloud backups.

If I’m going to be downvotes, could someone at least link to a document that shows how these statements are incorrect?

The statements only refer to:

“When files are created in Data Protection classes that aren’t accessible when the device is locked”

So not photos, most data.

This contradicts statements made by Apple, most information in iCloud backups is not encrypted end-to-end.

https://support.apple.com/en-us/HT202303

Apple, very clearly states that only the following data is end-to-end encrypted:

iCloud Keychain Payment information Wi-Fi network information Home data Siri information

So everything else, pictures, notes etc. etc. part of the iCloud backup, are not. Please not spread misinformation about this.

https://support.apple.com/en-us/HT202303

You’re just wrong, and spreading misinformation, please read the statements by Apple here:

https://support.apple.com/en-us/HT202303

In particular a limited subset of data uses end-to-end encryption, none of which is super interesting:

These features and their data are transmitted and stored in iCloud using end-to-end encryption: iCloud Keychain (Includes all of your saved accounts and passwords) Payment information Wi-Fi network information Home data Siri information

Unfortunately, we don’t understand the exact mechanism for many (most?) available medication.

Correlation != causation. Which is why you need to do randomized testing and try and account for other causative factors. But if you give a bunch of people a medication and they stop dying, you can pretty much say this is valuable and start giving it to people.

Right, but we’re talking about medical data. Which in lieu of further information I assume would be backed up in the same way as iCloud data. Unless there are statements to the contrary?

iCloud backups can be decrypted by Apple, because they have the key the data is encrypted with serverside. Are you saying this data is different?