Pages

September 08, 2006

Update on work towards $1000 genome sequencing

Currently it costs about $1,000,000 to sequence the genome.
A target is to sequence the genome for $1000 by 2014 (and $10,000 by 2009).

In 2005, Jay Shendure used off-the-shelf parts to determine the order of all the DNA bases in a bacterial genome, at 20 times the speed and one-ninth the cost of traditional DNA sequencing. Shendure is now building on polony sequencing, a method developed in George Church’s lab at Harvard. Shendure spreads millions of tiny beads on a glass slide, each attached to a small DNA fragment. He then adds fluorescently labeled DNA bases. The bases bind to short, complementary DNA sequences, and a standard fluorescence microscope records which base is at each position on a fragment.

Shendure next plans to use the technique to sequence the genome of a lung tumor in order to identify the genetic mutations that caused it.

Related articles:
Past review of costs for DNA sequencing

Commercial process for sequencing

The economist recently looked at synthetic biology It has a chart that looks at pyrosequencing (almost a billion sequences per person per day), ABI sequencers (About 5 million per person per day), EGEA gene writer (About 5 million per person per day) and ABI sythesizers (about 200,000 per person per day).

The article describes work to increase the number of amino acids that can be assembled into proteins. At the moment only 20 are used routinely in biology, but chemists can make thousands of others. Proteins containing those “non-biological” amino acids would have novel properties, and some of those properties might be useful. Lei Wang, of the Salk Institute in La Jolla, California, is trying to extend the amino-acid parts set. Dr Wang's starting point is the redundancy of the genetic code used by nucleic acids. This code is spelled out in the genetic “letters” A, C, G and T, which correspond to chemical sub-units of nucleic acids. The letters are grouped into three-letter “words” known as codons, meaning that there are 64 of them. All but three of the codons correspond to particular amino acids, and the order of the codons in the nucleic acid corresponds to the order of the amino acids in the protein. The remaining three are signals that the protein is complete.

Dr Wang has managed to reassign one of the stop codons in E. coli, the bacterial workhorse of geneticists, to recognise an unnatural amino acid. This can now be incorporated into proteins made by the bacterium.

September 07, 2006

Propellant free Microwave space propulsion

The EM drive engine takes microwave radiation produced by a magnetron and feeds it into a specially shaped resonant cavity. The waves push against the end wall. Because of the difference in wave velocity, being higher at one end that the other, there is a momentum transfer. The result is a measurable net force from the cavity against its surroundings.

The EmDrive is just entering the third of three development phases. The first was an experimental thruster, which Shawyer claims provided a thrust of 2g in over 240 tests, shown as a drop in the engine mass.

UPDATE: Mr Shawyer has already built a prototype emdrive capable of generating about 16 millinewtons of thrust using 1 kilowatt of electrical power. The next version could be several orders of magnitude more effective. (Supercooled and using Niobium) So several newtons per kilowatt? Current ion engines have 100 millinewtons for one kilowatt. Here is a pdf that describes improved ion drives and thrust to power for existing and proposed ion drives.

The second phase produced a demonstrator engine which has recently completed its own set of independently reviewed static thrust tests.

‘The next stage would ideally use technology employed by high-energy physics. If we approach a condition where we use the same shape cavity but at absolute zero and made from Niobium, the resistance drops to near zero and the Q value increases by several orders of magnitude.’

Shawyer cautions that the calculations only work for static thrusts. ‘You can’t beat the laws of physics. If it is used to accelerate, the Q value drops. It is best used to lift a body and oppose a force, for instance to counteract gravity. It cannot be used to accelerate further.


This would be great if it works. No moving parts. No propellant. Sounds like it would make it a lot cheaper to get to orbit. From orbit there are a lot of potential systems for moving quickly around the solar system. Plasma magnets, M2P2, ion drives etc...

Harvard University engineers demonstrate laser nanoantenna

It is called a plasmonic laser antenna, the design consists of a metallic nanostructure, known as an optical antenna, integrated onto the facet of a commercial semiconductor laser.


"The optical antenna collects light from the laser and concentrates it to an intense spot measuring tens of nanometers, or about one-thousandth the width of a single human hair," says Crozier. "The device could be integrated into optical data storage platforms and used to write bits far smaller than what's now possible with conventional methods. This could lead to vastly increased storage capacities in the terabyte range (a thousand gigabytes)."

The new device integrates an optical antenna and a laser into a single unit, consists of fewer components, has a smaller footprint (takes up less space), and benefits from an improved signal-to-noise ratio relative to previous approaches. The inventors expect, with further development, its wide adoption and use in academic and research settings as well as in the high-tech commercial sector.

"Eventually, we envision the laser integrated into new probes for biology like optical tweezers -- which can manipulate objects as small as a single atom," says Crozier. "It could also be used for integrated-circuit fabrication or to test impurities during the fabrication process itself. One day, consumers might be able to back up three terabytes data on one disk."

September 06, 2006

Bad predictions : failure to understand commercial

700 IEEE Fellows surveyed in a joint study by the Institute for the Future (IFTF) and IEEE Spectrum tried to get the IEEE Fellows to predict science and technology over next 10 to 50 years.

One of the questions and responses was:
Will a quantum computer reach the market?
Unlikely 42.7%
Equal chances 25.1%
Likely 22.1%

This is addressed at the dwave blog. Dwave is planning a commercial release in 2007. They note that a company is selling a NMR based quantum computer for $1 million. The NMR system has little power and is probably being bought for research purposes. If you make a $1 million sale then you have commerce and you have reached a market.

Many people who comment on the response at the dwave blog and perhaps some of the IEEE fellows were reading the question as "Will quantum computers exceed the capabilities of traditional computers and outsell them in the marketplace". That would be a higher hurdle. The first part I think will definitely happen as we get 100-200 qubit systems. The outselling portion is a far more uncertain situation and probably is unlikely to happen. A fairly ideal situation would be for robust quantum computers to become co-processors for special needs along with regular computer systems.

I think most of the questions in the survey are poorly phrased and some of responses probably indicate reading the questions with certain unstated assumptions.

I also think many of the IEEE fellows may not be up to speed on the actual leading edge of capability for many technologies. Otherwise they would not make the mistake of stating that something is unlikely to happen in 50 years when it has already happened.

Protein pathway to molecular nanotechnology

Jonathan Pugh asked about the protein pathway and how the recent breakthrough to apply variational analysis to allow for the calculation of protein folding a billion times simpler could advance molecular nanotechnology.

One of Eric Drexler earliest papers discussed using proteins for molecular nanotechnology.

A major milestone would be to make a version 2.0 of a ribosome
A ribosome can be thought of as a factory that builds a protein from a set of genetic instructions. A ribosome has 2.3 million base pairs.

This article points to surveys of DNA and protein related technologies that could lead to molecular nanotechnology

Freitas and Merkle has details on various approaches to positional assembly using Proteins

This was mentioned in an early post on this site

IBM to build Opteron-Cell hybrid supercomputer

IBM has won a bid to build a supercomputer called Roadrunner that will include not just conventional Opteron chips but also the Cell processor used in the Sony Playstation, CNET News.com has learned.

This computer when completed should have sustainable petaflop level performance.

Previous studies show that a modified cell chip could get 20 times the performance of an opteron

Recent nanotechnology related news

Applying variational principles may allow streamlining of protein folding calculations from trillions of steps to hundreds. A variational principle is a principle in physics which is expressed in terms of the calculus of variations. The new application of the old physics method is faster because it allows them to spend less time calculating motions of molecules stuck in quasi-stable intermediate steps along the folding process. The intermediate steps account for vast amounts of wasted computation time in traditional, step-by-step simulations. Determining the structure and folding of proteins is important to control the functions that they exhibit. Mastery of proteins could lead to molecular nanotechnology.


A pair of physicists at MIT have found a class of molecules that attach to nanotubes without damaging them. Instead of simply grabbing onto the nanotubes, the molecules latch onto the tubes and break some bonds in the nanotube walls. As a result, the sum total of bonds is the same before and after the handle is attached, keeping the nanotube's original characteristics relatively intact.

Researchers led by Kingo Uchida and Shinichiro Nakamura have now synthesized a compound in the diarylethene family whose surface becomes super-water-repellent on command.

Nanoscientists have transformed a molecule of chlorophyll-a from spinach into a complex biological switch that has possible future applications for green energy, technology and medicine.

Summary of some recent developments

Applying variational principles may allow streamlining of protein folding calculations from trillions of steps to hundreds. A variational principle is a principle in physics which is expressed in terms of the calculus of variations. The new application of the old physics method is faster because it allows them to spend less time calculating motions of molecules stuck in quasi-stable intermediate steps along the folding process. The intermediate steps account for vast amounts of wasted computation time in traditional, step-by-step simulations. Determining the structure and folding of proteins is important to control the functions that they exhibit. Mastery of proteins could lead to molecular nanotechnology.


A pair of physicists at MIT have found a class of molecules that attach to nanotubes without damaging them. Instead of simply grabbing onto the nanotubes, the molecules latch onto the tubes and break some bonds in the nanotube walls. As a result, the sum total of bonds is the same before and after the handle is attached, keeping the nanotube's original characteristics relatively intact.

Researchers led by Kingo Uchida and Shinichiro Nakamura have now synthesized a compound in the diarylethene family whose surface becomes super-water-repellent on command.