Skip to main content
Premium Trial:

Request an Annual Quote

Q&A: Henk Postma on Graphene Nanogaps for Electronic Single-Molecule Sequencing


HenkPostmaPhoto2_0.jpgName: Henk Postma
Age: 36
Position: Assistant professor, department of physics and astronomy, California State University Northridge, since 2006
Experience and Education:
Senior postdoctoral researcher, California Institute of Technology (worked with Marc Bockrath on carbon nanotube resonators, nano circuits, and nanorelays), 2004-2006
Postdoctoral researcher, California Institute of Technology (worked with Michael Roukes on high-frequency platinum nanowire NEMS resonators), 2002-2004
PhD in applied physics, Delft University of Technology, the Netherlands (worked with Cees Dekker on carbon nanotube junctions and devices), 2001
MS in physics, Leiden University, the Netherlands, 1997

Earlier this year, Henk Postma, a nanotechnology researcher at California State University Northridge, published an article in Nano Letters in which he proposed to sequence DNA using graphene nanogaps.

Graphene, a hexagonal carbon lattice that is just a single atom layer thick, was only discovered a few years ago and is becoming increasingly popular among researchers working on single-molecule nano-sequencing technologies because of its thinness, robustness, and good conductivity. Postma wants to run DNA molecules through a nanogap — a nanoscale slit in a graphene membrane — to read the transverse conductance of the DNA, which differs for each base.

At the Advanced DNA Sequencing Technology Development meeting of the National Human Genome Research Institute in Chapel Hill last week, In Sequence met with Postma to find out more about his approach to sequencing. Below is an edited version of the interview.

Can you briefly describe graphene nanogaps, and how you want to harness them for sequencing?

Graphene is a sheet of carbon that's a single atomic layer thick. It's a really fantastic material and was discovered only a few years ago. It's mechanically very robust, and it's actually also a really good conductor. I think it is a natural fit for doing transverse conductance sequencing of individual DNA molecules.

This is a very active field, and one of the really big challenges is, 'How do you electrically interrogate an individual base of DNA?' Some of this is material science — there was a talk [at the NHGRI conference] today about yttrium disilicide nanowires that are about 1 nanometer thick — but it's going to be really, really hard to scale that down so you can resolve the individual bases.

nanogap.jpgGraphene, because it is a single atomic layer thick material, is a natural choice for making electrodes for this kind of stuff. The other nice thing about it is, because it is mechanically so strong, you can use it as the membrane material as well as the electrode material. You automatically solve the problem that other people are having, who need to use electrodes to match up with the pore that they make in some other kind of material. So you basically kill two birds with one stone that way.

If you want to do this, you need really good control over the nanogap size — it has got to be down to 1 to 1.5 nanometers in size. I have done some work on what the signal would look like, because there is a challenge here. What we are trying to distinguish is individual nucleotides, by how their conductances are different. What if your nanogap is not 1.5, but 1.6 nanometers in size, or, what if there is a little step irregularity, so the DNA can go through the region where it's 1.5 nanometers, or it can go through the region where it's 1.6 nanometers? You are going to get different conductivity then. So how do you distinguish a current signal that varies because your nanogap varies from a signal that varies because the nucleotide is different?

[ pagebreak ]

What I put in this paper is that you can use a nonlinear current-voltage analysis to, basically, divide out the distance dependence. So you have a measure that is independent of the size of the nanogap.

I also did a few back-of-the-envelope calculations on the influence of noise. It turns out that electrical noise is actually rather insignificant. The first thing that's going to be a problem is, if your nanogap becomes too wide, the current peaks become very wide, and you start to get overlapping of adjacent nucleotide current signatures. I think that's the most dominant source of error. And, in fact, when I was doing the numerical analysis, up to about 1.5 or 1.6 nanometers, the error rate was numerically zero. So I think it's a very promising technique.

What is the main difference between this and nanopore approaches to sequencing single DNA molecules?

The solid-state nanopores require that you have some integrated electrodes. That's very challenging; getting those electrodes to align at the very end of your nanopore is really hard to do. Second of all, the electrodes need to be very thin, and that, I think, is a problem that has not been solved, unless there are reports that I am not aware of. If you use graphene, you don't have to worry about this stuff, you get the ability to distinguish individual nucleotides for free. The other difference is the fact that because the membrane is the electrode material, you need to use a nanogap, versus a nanopore.

You can compare the approach to solid-state nanopores in terms of its robustness, how stable it is, how strong it is, how well it can deal with pH and temperature fluctuations, and all those kinds of things, but it is also very similar to the biological nanopores in size, because it is very shallow. If you compare this to the alpha-hemolysin nanopore, the narrowest point is a fraction of a nanometer deep, so it's very comparable to the graphene thickness.

[A graphene nanogap] would measure DNA at one point, with all the attributes of nanopore sequencing, so presumably really long read lengths are possible, and you can distinguish repeat-rich regions because you can sequence those reliably — things that are typically very hard to do right now with current sequencing technology.

What is going to be the greatest challenge for building this?

At this point, it's making the nanogaps small enough. We have about four different technologies that we have identified, and we are working on all four. We have got it down to about 8 nanometers right now, and we have some ideas of how to get it to become smaller, but I don't want to talk about that right now. We need to get it to 1.5 nanometers. That's not so bad — we started at 20 nanometers, so if we do one order of magnitude per year, I think we will be there next year.

How long will it take to develop this?

It's going to take a couple of years. First of all, there is the nanogap fabrication, then there is the whole integration, then it is doing the electrical transverse measurements at really high speed, which is not trivial. We have the capability to do high-impedance measurements at really high frequency; we just finished that a couple of weeks ago. But it's far from trivial, and then we have to integrate that into a nanofluidic setup with this device in there. So there is a lot of integration that needs to happen. But I think the first thing — and that's what we are focusing our effort on — is making nanogaps that are small enough.

What are possible limitations of this approach?

The graphene has this hexagonal lattice that likes to stick to DNA. So you need to actually put something on it to prevent the DNA from sticking to it. At this point, that's the only challenge I see. But we have some ideas on how to do that, it's not that hard.

How easy or difficult will it be to scale this up?

What we are doing is single-device proof-of-principle. It depends on how easy it ends up being making these nanogaps with the right length. All the other things are relatively scalable. But the nice thing is that you don't actually need a lot of parallelism because you have such a long read length. So you might be able to get away with just a few devices that you can use to sequence a whole human genome.

[ pagebreak ]

So in theory, the read length would be unlimited?

I'm not sure what 'unlimited' means because I think there are real limits to it. I'm cautiously optimistic that we will be able to be reach more than a million bases, but I don't know for sure until we actually do it.

Do you already have plans for how to commercialize this?

I think it's too early for commercialization. Of course we are thinking about it, but no comment.

Have you filed IP on this?


Are you planning to do so?

No comment.

Do you have any competitors?

Not on the graphene nanogaps. I think people are warming up to the idea of using graphene, which is, I think, a natural choice. I think the time is right for using graphene. People are using graphene nanopores, and I'm really interested to see what they are doing with that. All this stuff is unpublished, new work.

How did you come to work on graphene nanogap sequencing?

I have a background in nanoelectronics. When I was doing my PhD in Delft, I studied carbon nanotubes, so I have always been very interested in hexagonal carbon lattices, except these were rolled up pieces of graphene. And then I went to Caltech and I started doing really high-frequency measurements of mechanical motion of these carbon nanotubes and nanowires.

I had been thinking about a dielectric measurement to do sequencing since I started at Caltech in 2002. Then I saw a paper on graphene in 2004 and started thinking about using that for dielectric sequencing. The paper that put me on the trail of conductance measurements was published by Michael Zwolak in 2005, talking about the transverse conductivity of DNA. In 2006, I started my own lab, and we have been working on this since then, basically.

The Scan

Shape of Them All

According to BBC News, researchers have developed a protein structure database that includes much of the human proteome.

For Flu and More

The Wall Street Journal reports that several vaccine developers are working on mRNA-based vaccines for influenza.

To Boost Women

China's Ministry of Science and Technology aims to boost the number of female researchers through a new policy, reports the South China Morning Post.

Science Papers Describe Approach to Predict Chemotherapeutic Response, Role of Transcriptional Noise

In Science this week: neural network to predict chemotherapeutic response in cancer patients, and more.