bioinformatics, genomes, biology etc. "I don't mean to sound angry and cynical, but I am, so that's how it comes across"

We have the $1000 genome; what’s next?

Well, we got there, didn’t we?  And when I say “we”, I mean “Illumina”.  The $1000 genome is clearly here.  This has been a goal of genomics for so long, that we are left asking – what’s next?  If any of you are thinking “The $100 genome!” then please leave this blog now – you’re not welcome.  Obvious changes like this are intellectually bankrupt, and annoy the hell out of me.

The next step is pretty obvious, and I won’t be the first to say this: “Genome at home“.  That should be the next challenge of genomics, the equivalent of “The $1000 genome”.  And when I say “Genome at home”, I mean everything at home – sequencing and analysis.  What we need is technology that can take a sample from a person sitting in their own home, sequence the genome, and upload the data to software sitting on a laptop that can analyse the data and tell the person what it means.

I can already anticipate the comments/emails from companies telling me they can already do this (at least from a software perspective).  Save your “ink” – you can’t.  Keep trying though.

Some people may point towards the MinION USB sequencer, and I thnk this is the closest device to being able to generate a “Genome at home”, but there are three barriers still to be overcome: 1) I don’t think the MinION throughput is human-genome ready yet; 2) sample prep still needs to be done, and you need molecular biology skills to do it; 3) we don’t know how good the data are yet.

Of course, as is true of every technology, the “raw data to clinical interpretation” software doesn’t exist yet, though many are trying.

So there we are – the challenge that I think should replace “The $1000 genome” is “Genome at home“.


  1. This may be controversial, but how about this as a successor to the $1,000 genome…
    …the *complete* and *accurate* genome.
    I know, crazy idea. It will never catch on.

    • Do we need a complete genome for clinical interpretation? We won’t find all disease variants, but we will surely find enough to make it worthwhile?

      • Better to have a single molecule analysis so without the bias of current technologies……..and how about direct detection of the methyl C and even the other C derivates and beyond.
        The epigenome is the next goal.

      • Apologies I should have written the practically infinite epigenome.

      • For clinical applications, a completely assembled genome mostly isn’t really necessary, since analysis would be done in respect to reference genomes. But still, sequencing costs, and flexibility thereof will determine the viability of any such application.

  2. I’m not convinced the “genome at home” goal is one that anyone is particularly interested in pursuing, except a few hobbyists. “Genome in the clinic” is a closer goal, and it will require some pretty major improvements in data handling, in privacy laws, and in interpretation of genomic data. Faster turnaround might be important for that also.

    The $1000 genome isn’t really here anyway—that figure is only for the sequencing costs of redoing a human genome, not sample prep and not the analysis (which is still quite expensive, even for ones that are clinically relevant, like cancer tumor genomes)—and de novo animal and plant genomes still cost a LOT more than that. There’s also still been fairly little progress on the centromere and telomere regions, which have some interesting genes buried in a lot of hard-to-sequence repeats.

    With PacBio sequencing we are getting close to the $1000 bacterial genome, though error rates are still a bit too high unless short read data is collected as well. I’m really curious whether the MinION will finally this year prove to be a viable alternative to PacBio—it would be good to have another long-read technology available, with a different error pattern.

    • Whilst I agree that clinical genomics is a challenge, it is no longer a technical challenge, more a challenge of politics and opinions. The $1000 genome was originally a technological challenge, and I think the “genome at home” is a similar challenge. I am very surprised to hear you say that it would only ever be for hobbyists! I think if it was cheap and reliable enough, everyone would do it – call it the 7 billion genomes project if you like 😉

      The $1000 genome *does* include sample prep, but you are correct, the interpretation is key – I anticipate that commercial vendors will meet this challenge in the next 2-3 years – desktop software that will take fastq and produce clinical interpretation of variants.

  3. Clive G. Brown

    2nd June 2014 at 4:54 pm

    I’m not sure its your genome that you’d want to do at home, but an inventory of the free floating nucleic acids in your blood.

    • Cell-free DNA would be quite interesting (for cancer monitoring). I also think transcriptome monitoring will end up being interesting for general health monitoring.

      • Clive G. Brown

        3rd June 2014 at 7:18 am

        yes that sort of thing, streamed to a service, trending over time like your Fuel//Up, crowd sourced interpretation.

  4. The molecular biology skills required for the MinION sample prep are simple enough you should be able to teach most middle/high school students to do it. That’s really not a barrier.

    • Clive G. Brown

      3rd June 2014 at 1:20 pm

      always working on something a lot simpler …

    • Yeah, I think what I’m really talking about here, if you like, is “pee-on-a-stick” genome sequencing. Something that can be done in a kitchen, bathroom, bedroom, hospital, field, barn etc.

      • It’s not hard to imagine the path to that (look at the clever one tube NEBNext kits that use temperature to activate/inactivate various steps of a reaction) — but remember it took a LONG time to go from the first home pregnancy kits to the state of the art (3 decades?) — the early ones involved multiple steps & looking during a specific time window to get correct results. I do already believe you could carry a complete MinION arrangement in a small daypack — laptop with cellular connection, MinION, small cooler for reagents & a portable drill with appropriate attachment playing centrifuge.

  5. I think going down to “$100 per Genome” will both be a significant challenge and a significant benefit.

    For example, at $100 you can routinely use sequencing for animal breeding and conservation programs, at $1000 you can’t. Especially if it takes serious time on 1GB RAM machines to assemble the genome. (Though we can call variants easier, but anyway…).

    Also, routine antibiograms or other diagnostics through sequencing at $100 are feasible, at $1000 it’s not.

    The killer feature of the nanopore in these settings will be read length, since the price is roughly ten times higher than with other technologies. For DNA Barcoding, you could just sequence the complete mitochondrial DNA, instead of just selecting a gene fragment or two.

    The MinION could be used in the field to sequence a mixture of larvae from a pond to assess biodiversity or guide the collection process.

    For “Genome at home” even $100 wouldn’t be cheap enough to compete with service providers, and the immediate applications would be rather limited. Let’s solve some urgent biological problems and find a few block buster drugs first, ok?

  6. Genome at Home. Needs ( only) one (decent) application, robust, with life changing consequences. For enough people. Then done. Think home pregnancy test.

    • Clive G. Brown

      4th June 2014 at 8:59 am

      Exactly, but i don’t think its ‘whole genome’ sequencing as such – instead it is done by picking up circulating nucleic acids (whether DNA/RNA or mutant, modified, viral/bacterial, chimeric, randomly sampled or selected for, counted). It may not be limited to one-off present/absent testing either (although this will be important), I think perhaps more continuous monitoring and trending.

Leave a Reply

Your email address will not be published.




© 2017 Opiniomics

Theme by Anders NorenUp ↑