Sequence DNA on your laptop (or in space)

[Bob Eisenman]

 

Sequence DNA on a USB device for about $1,000

Around 1995 I was working for a DNA sequencing related project in Boston. Groups around the country were working independently as in some on chromosome 1 stuff and another on chr X and another redundantly working on the same chromosome region  Medical data strongly supports the correlation between specific genes and certain medical conditions. In the 1990's the methods required to sequence DNA used radioactive isotopes of either sulfur or phosphorus and ran overnight 'gels' about the size of a page from the NY Times. The next day an autorad was made by placing film on top of the dried gel affixed to chromatography paper, the waste isotopes were poured down the drain (according to legal limits) and after a period of a few hours a set of hand recorded bases numbering about 150 total was recorded. 

After congressional funding was obtained new techniques were developed to sequence DNA in reads as long as several hundred bases per night without using isotopes. The entire genome of humans was finished and made available to the public around 2000.

Between then and now a new sequencing method called 'next generation sequencing' became available whereby an entire human genome could be sequenced in a matter of days using 'massively parallel' computer processing.

In the last 5 to 10 years a new sequencing method based entirely on microfluidics and the electronic detection of the bases of a single strand of DNA passing through a small microfluidics system micropore became a reality. The first microfluidics DNA sequencers were erroneous at rates up to 30%. More recently that error rate has been reduced significantly. 

I was working one night on my laptop when a LinkedIn notification announced that a fellow I did some lab work for had been at his new company for a full year. Looking at his LinkedIn posts turned up a picture of him standing next to space shuttle astronaut 'Kate Rubins'. She used a microfluidics based DNA sequencer on the space shuttle. 

The 'Minion microfluidics' DNA sequencer fits onto a USB sized device and plugs into a laptop via a USB 3 port. The cost of a device of this type is about $1,000. The company URL is:

https://nanoporetech.com/products/minion

 

[Hunka Hunka Burning Love]

Would this device allow you to do paternity tests on the go? Just curious, that's all.  

BTW I think that lady on the right looks a little like a dude.

 

 

[ramma]

I remember learning about this device part way through my NGS work. Really neat device, but as far as high depth and accuracy research is concerned it has a ways to go. It's newest revision does 10 Gb where as the Illumina NGS systems I worked on are between 300-1000 Gb, plus have a higher accuracy in most situations. The big benefit to such huge output is being able to put multiple samples in the same run, since at the time we were still spending $4-10k per chip with the tagging of samples being virtually free if done in our lab. That was for a research center owned machine even! To cut costs we'd end up waiting a week or two for other research center labs to have sequencing samples prepped. If only the chinese illumina farms turnaround had been faster since they dropped below $1000 per full genome runs early on in my bioninformatics adventures.

I should keep more up to date on NGS. The Pacific Bioscience real time sequencing was what I heard about most, but honestly the stats don't look that amazing. In reality real time data isn't that useful either.

[Bob Eisenman]

https://en.m.wikipedia.org/wiki/Kathleen_Rubins

Dr. Rubins looks about the same as her Wikipedia photo.

On 5/4/2017 at 9:12 PM, Hunka Hunka Burning Love said:

Would this device allow you to do paternity tests on the go? Just curious, that's all.  

BTW I think that lady on the right looks a little like a dude.

 

Paternity tests... Humm...Maybe 'ramma' would be a better source of authority. 

[Bob Eisenman]

On 5/4/2017 at 9:12 PM, Hunka Hunka Burning Love said:

BTW I think that lady on the right looks a little like a dude.

Maybe if she cut her hair and took guitar lessons she would look and seem more like Jess Lewis

 

[Bob Eisenman]

On 5/5/2017 at 3:11 AM, ramma said:

$1000 per full genome

https://www.illumina.com/company/news-center/feature-articles/illumina-scales-for-growth-in-china.html

 

"What applications are really growing right now?

The first application that comes to mind is population-scale sequencing programs and services. Illumina’s HiSeq X Ten, which enabled the $1,000 genome, has been strongly adopted inside China. "

[Bob Eisenman]

On ‎5‎/‎4‎/‎2017 at 9:12 PM, Hunka Hunka Burning Love said:

Would this device allow you to do paternity tests on the go? Just curious, that's all

Maybe this YouTube video gets closer to an answer to your question. SNP (single nucleotide polymorphisms) identification  are places where your dna differs from  someone else.

https://www.youtube.com/watch?v=wgtmA88XYxQ

 

[Bob Eisenman]

On 5/4/2017 at 9:12 PM, Hunka Hunka Burning Love said:

Would this device allow you to do paternity tests on the go?

Probably not in its current state (paternity and forensic ID differ) and sample prep method(mechanical shearing) but according to

https://www.nature.com/articles/srep41759

"This study demonstrates proof-of-concept forensic SNP genotyping using the Oxford Nanopore MinION sequencing platform and shows the current capabilities of the system. "

"All but one of the 52 loci were genotyped correctly. We identified two SNP loci that prove to be problematic to genotype robustly using nanopore sequencing."

"When these loci are avoided, correct forensic genotyping using nanopore sequencing is technically feasible. "

"The technique is however still subpar compared with current techniques such as capillary electrophoresis and Illumina sequencing in terms of costs, analysis time, sequence error rate, representation bias and allelic imbalance. With the ongoing improvements, nanopore sequencing may become suitable for routine use in the future."

Maybe a pcr based DNA sample amplification (requires a lab, pcr primers, tech and whatever software they use to work out DNA based paternity IDs) of a common paternal genomic SNP testing region could be used. 

For $1000 a person a whole genome could be obtained by better methods.l

[Bob Eisenman]

On 5/4/2017 at 9:12 PM, Hunka Hunka Burning Love said:

Would this device allow you to do paternity tests on the go?

For the real deal try ($80) (Google search hit)

http://www.swabtest.com

[ramma]

5 hours ago, Bob Eisenman said:

For $1000 a person a whole genome could be obtained by better methods.

Yep, full genome is most accurate, but while $1000 is cheap as fuck for most applications it's insanely expensive for everyday paternity testing (at least out of the US).

Paternity tests are as cheap as they are because a limited number of known sections of DNA called microsattelites is analysed, not the whole genome. Those short sequences represent potential allele inheritance from the parents and are amplified (copied), read, then compared between the parent and childs samples. The results represent a % of matching reads. It's basically the same as what personal sequencing products like 23andme do, but with less a lot less reads.

One interesting aspect of those tests is that they can't easily account for conditions like chimerism, which is where an organism has cells with varying alleles throughout their body. It confuses a lot of tests since the different cells tend to stay in groups as they divide, so a DNA sample could vary from one area to another, or be a mix of the two distinct genomes. It can also create weird situations like animals/people with two blood types.

[ramma]

If you get really bored, here's some videos showing exactly how a paternity test works (DNA fingerprinting), and how a full genome sequencing method works (Illumina, since that's all I've worked with). I can't find a good video of how computer analysis of DNA fingerprinting works as opposed to the gel electrophoresis method, but the technique up till then and the end results are really similar (the computer is more precise, but costs more).

 

[Bob Eisenman]

 

11 hours ago, ramma said:

If you get really bored

Thanks, I'll have to watch it. I haven't been a part of the sequencing scene since 2006. The core sequencing facility went private later and became Claritas Genomics (ask for Hal Schneider)

http://claritasgenomics.com/ordering/tests/all

https://www.linkedin.com/in/hal-schneider-6ab6722a/

I'm about as sequencing literate as using 'Mutation Surveyor or picking a PCR

primer from scratch using the rules:

   aim for an 18 to 21 mer with about 50 % GC

   no more than three or four consecutive bases of the same call

   no hairpins

   end the primer on each end with a G or C to hold it down

or just use a Primer3 website

http://www.simgene.com/Primer3

There is an interesting isothermal amplification company parented off New England Biolabs called 'Biohelix'. Their kits use a helicase and isothermal amplification system. They weren't interested in hiring another tech in 2007.