The sequencing... to determine the sequence of letters of building blocks in a nucleic acid is a very important problem nowadays. For instance for the human genome project. Now the difficulty is that you usually have many such molecules. If you degrade them, you have to do that step by step because of the stochasticity of the reaction. One molecule might go faster, the other goes a little slower, so if you want to see the sequence letter by letter by letter you have to do the first letter of all of them, then you have to stop the reaction. Then have to do the second letter, and that takes time. So a sequence velocity nowadays reaches in the best case some one thousand per day or a little more, and you have to... you can easily calculate what you have to do if you want to sequence one human genome. A human genome has three billion such nucleotides, and if you can do only a thousand per day you need three million days... that's impossible. Of course you can use many machines in parallel as people do, and you can use some tricks perhaps to get a little faster, but still with the present technology human genome project is a...

[Q] Very time consuming.

... time consuming, very involved procedure. Now, why not using again nature's technology? Use an enzyme which degrades the nucleic acid from the end, that's called an exonuclease, and pick up the single monomers when they appear in your focus of the... Because we have a method by which we can pick up single molecules and you know you can do that only with single molecules, because of the stochastic events and it would... If you would have ten molecules the first one would appear then, then the second one of the other molecule appears earlier than the other, so there soon would be a mesh of reactions. But with the single molecule they come step by step as they are degraded by the molecule, and now you can calculate.  The enzyme does a job in a hundredth of a second, and the day has 10⁵ seconds, a hundred thousand seconds, so the principal limit would be ten million per day, which would mean in a hundred days you can do a human genome.

So we are working on this together with other groups because this is a high-tech problem in nanotechnology to machine those things, and we are working with people who produce these devices in nanotechnology, that's one possible application. But I mentioned it in here because that's where the limit is if you do evolutionary experiments, if you do... You have to come down to the sizes which nowadays are accessible to nanotechnology, you have to use analytical methods which can... that means you have to use fluctuation, correlation methods. But then you really reach a range in which you, well... in which you do molecular...

[Q] Single molecular...

... analysis. Single molecular. And now the consequences is not only biotechnology, you can develop now a molecular diagnostics for medicine... in other words you can really look at single virus particles, you can look at prions and, oh, lots of new words. So that's indeed something we are also very interested in and we do work on.