New chip delivers DNA results within an hour | Hacker News Reader

New chip delivers DNA results within an hour

dangerousprototypes.com

70 pointsbcl20 comments

Note that this chip isn't doing genome sequencing - it's looking for SNPs, i.e. a bunch of known single letter changes in the DNA. This chip chops up the DNA, runs high-speed PCR to amplify the segments, filters the DNA, and then uses a new electrochemical sensor to detect the SNPs.

There's considerably more information at http://www.diginfo.tv/v/13-0022-r-en.php

Yeah, I was expecting whole exome in a few hours or something. I'm waiting for that press release still.

Exactly. It's not clear to me whether they're only checking for SNPs in a single locus or if they can multiplex (or do whole exome). Though Oxford NanoTech is still essentially vaporware at this point, their technology excites me a lot more than something like this.

Yes, and before them PACB's vaporware excited me. Lets just be happy we can still get excited about something. :>

On the whole, this "announcement" doesn't really provide a lot of detail, and some of the scientific explanation is misleading (or just wrong). That doesn't mean the development isn't exciting - doing a PCR on a chip that size is pretty cool.

I am going to guess that the chip is similar in design to that of Ion Torrent's DNA sequencer. I believe that Ion Torrent's DNA sequencer is nothing but millions of little pH sensors.

In any case, THIS IS F*ING COOL!

I used to work for Ion in R&D for their sequencers. Sales would claim, "the chip is the machine!" However, the chip really does nothing in Ion's case. Like you said, millions of little ph sensors. This thing actually lives up to the Ion Torrent promise. They could gain a lot of traction if they actually marketed it that way.

> This thing actually lives up to the Ion Torrent promise.

Are you sure of this? The video makes it sound like this thing is a sort of automated SNP chip, which is great, but is far less than what Ion Torrent is promising. You're getting a list of SNPs at known sites--no data on rare or novel variants. No sequence data also means you are powerless to detect any sort of structural variation.

That was my take as well, but the whole thing is so vague that it's hard to be sure.

Those are valid points, and we can go further and debate how suitable Ion is as a platform even for novel and structural variants. However, in terms of what Ion promised early on, and the talks they gave at genetic sequencing conferences -- the chip is far from the complete machine. Maybe I'm getting a little too semantical here, but this panasonic device does everything on the chip.

Correct - Ion Torrent (and Oxford Nanopore) promised a `sequencer on a chip` and this isn't that, it's `PCR on a chip`.

That said, PCR for known SNPs (breast cancer, alzheima's etc) in this timescale could revolutionise modern medicine (no hyperbole!)

I'm curious what this will do to disrupt the efforts of companies offering tests using the current array of DNA sequencing technologies (which are probably mostly Illumina Hi-Seqs and MiSeqs).

This piece of equipment seems like it cuts out a lot of the pain of those technologies: massive data output, requirement of expensive compute resources, a team of bioinformaticians, etc.

Not much in the short term. This sort of thing still isn't really suited to the sort of industrialized testing most commercial labs are undertaking. Maybe when the price comes down enough or they're reusable in arrays it might have more of an effect, but at the moment it's still more reasonable to hire a few entry-level people to perform DNA extraction and PCR.

I think this cuts out the pain more on the lab side. Note that this is a SNP array, not a sequencer. Current-gen SNP array output is already relatively painless to work with (depending on manufacturer, I assume - Affymetrix 6.0 arrays were pretty easy).

But this technology appears to automate a lot of the lab side: restriction enzymes to chop it up, PCR to amplify, etc., which is also applicable to sequencing.

As far as I could understand it's only finding common SNPs, so it will do nothing really to compete with Hiseq or Miseq which are doing a compeletely different thing. However I find the technology inside pretty fascinating nonetheless - 30 PCR cycles in 9 minutes all on a chip smaller than a business card seems extremely cool to me (but then I'm a bioinformatician - I get excited when I see a pipette).

"Listen punk, we have your DNA and we'll have the results in under an hour. Why don't you just come clean now and we can make this a lot easier on everyone."

This law and order episode is being written as we speak.

What's the cost going to be like? Is it cheap enough that I can do DNA fingerprinting through SNP matching at home? If so it could be a fun project.

Man, those 90° traces are uuuu-gly!

Still pretty darn cool though.

One step closer to a portable midi-chlorian detector.

Reply on news.ycombinator.com