Nanopore DNA Sequencing

mindpixel writes: "Harvard scientists have concieved a revolutionary technology for probing, and eventually sequencing, individual DNA molecules using single-channel recording techniques. The technique essentially pulls a single strand of DNA through a nanopore, reading off the individual bases electrically. The technique could allow for decoding of a person's genome in hours instead of years." While the sequencing in hours instead of years is something that's pretty darn cool, our holdup in using this data is actually now what the genes are, and how they interact. That will still take years for us to figure out.

Re:After ten years sequencencing the human genome

More often than not, doing things the "hard way" is an essential first step in figuring out how to do it the "easy way". I can think of innumerable examples where people have worked hard to accomplish something significant only to have someone come out with an easier way to do it. But this is crucial to the development of science and technology. Take for example PCR used in genetic analysis. One of the guys down the hall used to spend hours and hours and hours leading to days just getting the reaction products to the point where he can analyze them. Now we can accomplish this in just a couple of hours. Or what about something a little closer to home for the Slashdot crowd? I can remember my father having to program the mainframe system at the university using paper tape and punch cards. God forbid if you got anything out of order or misplaced a few paper punches. There was a logic to programming those things, but trouble shooting could be much more complex than it is now with automated trouble screening for code in development applications. Additionally, the same program my father took several weeks to write I can now write in a couple of hours with current development tools and I get visualization tools that could not even be dreamed of when my father was in graduate school. Additionally, we have the luxury of unbelievable speed of analysis on a Powerbook I can tote around in a backpack rather than trying to get time on the campus mainframe at 3:00 in the morning.

My point is that all this pain grief and expense was absolutely required to get where we are today and in most fields this will continue to be the way things are as we have no way of knowing what is hard and how it can be easy if what we are doing is new and innovative.

A genome a day....

Firstly, this idea has been around a while (i.e. check the references on the article), but it does appear to be getting there now. Perhaps 5-10 years before we start seeing commercial sequencing machines based on this technology???

It will not replace conventional sequencing technology, unless it can beat the now pretty cheap cost. Conventional sequencing is based on labelling the individual DNA bases with a different flourescent dye, and running the DNA through a gel which seperates the DNA according to size: As each base runs through the gel, it goes past a laser/detector which can detect the specific DNA base (A,T,C or G) at that position. Due to gradual impovements to this technique over the last 20 or so years (originally it employed radiation, rather then flourescence) the speed, sensitivity and cost has decreased dramatically. For example, the human genome project started in ernest about 10 yrs ago. Celera Genomics, using modern technology (and alot of financial backing, and the fact they are a subsidiary of the people who make sequencing machines, [appliedbiosystems.com]) competed the genome in a matter of months. The increase in DNA sequencing capacity puts moore's law to shame.

For example, our lab could process around 100kb (thats KiloBases guys!) of data a day, but we never even touch this with our machine [appliedbiosystems.com]. No need, and the same stands for many small-medium research labs. Alot of people like us will probably stick with conventional sequencing technology for a long time (it works well, is high enough throuput, cheap & easy).

However, the are some exciting applications with single strand sequencing. For example forensics. Also, it allows the oppotunirty of sequencing RNA (this is the "messenger" which passes the "important" part of the DNA message to the ribosomes, which then "compile" a protein - the stuff which actually does things, like an enzyme or structural component). Sequencing RNA is exciting, as currently you have to convert the RNA back to DNA (which can cause problems) and then sequence that.

Another obvious application for this would be very high throuput sequencing which would be employed by the major sequencing centres. Yes, i know we already have the Human Genome, but a fashionable idea at the moment is comparative genomics. This is very much taking biology back to its roots (i.e. like Darwin and Wallace comparing the morphological characteristics of certain species and infering adaption), but at a molecular level. This will yield amazing insights with discoveries having important implications from medicine to evolution. In fact I think the general public & media will soon be bored of this. Each week it will be a new genome being announced; mouse, chicken, rat, pufferfish, rice, corn, dog, cat, cow, chimp......

Re:Huge!

without the mess of Gel Electrophoresis would be HUGE

Hey, dont knock electrophoresis mate. It is the basis of PCR resolution, Southerns, Northens, SSCP, conventional sequencing methods plus a multitude of other applications. Furthermore, refinements to this approach (read: capillary electrophoresis) have supplied one of the major advances to sequencing methodology in recent years, unlike the technology we are (were?) discussing.

Who's GNOME did we sequence anyway

Good Question. Apparantly, they took DNA from around 100 (i forget the exact figures here) US citizens of various sex/ethnicity, picked 7 out of the hat, and sequenced portions of each (most from a single, unidentified individual - although if nanopore technology comes to fruition, i reckons we can track him) down

p.s. thats one of my favourite all-time techno-typo's: the Human Gnome Project. Almost as good as sequencing my ARS (a yeast thang....)

DNA Code

We now get to the point of where this is sort of link looking at the source of the Linux kernel, if it all was written entirely in Finnish [clinet.fi] in the first place, and we knew almost nothing about Finnish [finland.fi] (which is not an IndoEuropean Language [malibutelecom.fi]) and almost nothing about any kind of programming.

But it is a place to start.

Side note:

while looking up the Finnish Language pages for this comment, I came across this tidbit: That Finnish has "no equivalent of the verb to have [finland.fi]". This has interesting philosophic implications in the history of open source, etc.

Check out the Vinny the Vampire [eplugz.com] comic strip

Re:A genome a day....

Also, DNA breaks very easily. No way are you going to be able to pull a whole chromosome through at once

Yeah. Sure, i agree with you. Once youve taken DNA out of the nucleus and stripped out the protein complement of chromatin, it is darn hard to get high-quality genomic DNA (for PFGE or whatever). Furthermore, yanking the DNA through that little hole is gonna probably cause problems.

However, you dont need unfragmented DNA. For example, you could fragment genomic DNA, and pieces of this would randomly pass through the pore - essentially a shotgun approach. This wouldn't be a bad way of doing it. This would also get round problems with the detection mechanism. For example, if they could only disringuish between C/T with 80% accuracy, multiple reads of the same sequence could clear this up.

The technology could also overcome problems such as cloning bias, problems with sequencing microsats (e.g. (AT)n), GC rich regions etc. Which the HGMP is still having problems with.

Sounds like good news for systematic zoology!

I am a zoological systematist, working in entomology.

The human genomes was sequenced by taking lots of DNA, cutting it up randomly sequencing the random pieces of cut up DNA.

In my field, we work with much smaller amounts of DNA. Sometimes I only have a single specimen of a tiny insect, or unique material (from rare or extinct species) to try and get some DNA out of. In older material, DNA is usually degraded and many times we end up with nothing but a destroyed or damaged specimen.

With small amounts of DNA to begin with, we have to amplify (PCR) single genes or regions by using general primers, which means that they don't only fit on the insect DNA, but fungi and human DNA too, making contamination of your material very real risk.

If this technology turns out to work on a larger scale, it's amazing news for me and my collegues.

The nanopore technolgy sequences single moleculer, which means the PCR step becomes unneccesary! This means that we can get sequences from specimens with severely degrades DNA, and we don't have to be as afraid of grinding up rare material in hope of getting sequences.