Reading genomes for ridiculous money
Genome reading has fallen to $4,400
Artem Tuntsov, Infox.ruA year ago, the hitherto unknown Californian company Complete Genomics shocked the public by promising to read the human genome at a price of $5 thousand in the first half of 2009.
The audacity at that time was unheard of – the cost of the cheapest genome at that time was at the level of $ 70-100 thousand. So much were the consumables required to decode the genome of a man from the Nigerian Yoruba tribe, which competitors from Applied Biosystems completed in February 2008.
Faster than a computerMeanwhile, if we are guided by the genetic analogue of Moore's law, the promise of Complete Genomics was quite natural – moreover, whoever was the first to believe in it, he should have been on the horse, having collected the maximum orders.
Moore's law, as you know, states that the performance of computers doubles every year and a half. If you look at the price of reading the human genome, it fell by an average of ten times over the same period.
A year before the Nigerian, the discoverer of its double structure, James Watson, found out his DNA sequence, which cost about $ 1 million, and a year and a half before that, the DNA of his eternal antagonist Craig Venter was read, which cost about $ 10 million. The cost of the reference human genome, the synthesis of which was completed in 2003, was about $ 300-400 million; however, the latter amount includes all costs, and it is incorrect to compare it with the cost of consumables alone (as for the Nigerian, Watson and Venter).
Saving on everything
The latest issue of Science published an article by scientists from Complete Genomics and the medical faculties of Harvard and Washington University in St. Louis, which proves that the company's promises were not in vain. The authors of the work under the guidance of the chief scientist of Complete Genomics (there is a position with such a funny name in many American companies) Drmanacha's radio describes the reading of three complete human genomes, which cost from $1,726 to $8005. The average cost was $4,400. The new VAZ-2105, for comparison, costs about $5,600 today.
The key to lower prices is lower reagent consumption. Numerous enzymes necessary for cutting and multiplying DNA, as well as reading the obtained fragments, are very expensive, and it is necessary to wash samples of the studied molecule with them many times. To avoid this, the company Helicos, which had a record of cheapness for the last three months, does not reproduce DNA at all, but reads a single molecule cut into small pieces. The cost of reagents required for the August sequencing by this method was $ 50 thousand.
Complete Genomics went the other way. The new technology also requires DNA reproduction, but this is done in a rather tricky way. Scientists have tried to miniaturize all chemical manipulations so that expensive substances do not spread over glass plates for nothing. In addition, the reproduction process itself is arranged so that each enzyme molecule performs its work several dozen times.
Nanosheets
It all starts with the fact that the desired DNA molecule is broken into short fragments using ultrasound. Then the ends of these fragments are glued together, and pre-known, "marker" oligonucleotides of strictly defined length are inserted into the resulting ring in the right places. And then, probably, the most interesting trick is that scientists add a special enzyme to these DNA rings, which some viruses use to copy their ring–shaped DNA. This enzyme works non-stop, circling around the ring until it is washed away or there are no nucleotides in the vicinity of the enzyme from which copies can be made.
As a result, the enzyme is followed by a long tail of numerous copies of the ring (together with marker oligonucleotides embedded in it), which curls up by itself into a ball, somewhat resembling a tangled fishing line - only smaller than a micron in size. Scientists call these lumps DNA nanosheets, and it is from them that the sequence of DNA nucleotides is eventually read.
Scientists place these lumps on a thin plate measuring 25 mm by 75 mm, in which 350 million microscopic depressions are lined up in straight rows, shoulder to shoulder (so as not to waste reagents in vain). Once in the recess, the DNA nanoparticle is retained there by intermolecular forces, so that as a result, 350 million samples are obtained on each plate, each of which contains several dozen nucleotides of the original DNA – repeatedly multiplied and tangled into a lump. It remains only to count them.
And so it will do
Scientists use light to read. Special glowing indicator molecules are mixed with DNA arrays. They are securely attached to the DNA under study only if one of the ends of the molecule is held by the markers inserted into the original ring, and the second by a certain DNA nucleotide located at a strictly defined distance from the end of the marker. Such indicators are added in four variants, glowing in different colors, depending on the nucleotide with which they mate.
All the extra indicators that are not attached to the DNA are washed away, so that the color of the glow immediately shows which nucleotide is in a given position. This procedure is repeated several times, each time moving another step away from the marker area. This is how scientists learn the entire sequence between markers. Then it remains only to link these short sequences together, comparing them with the reference genome. The whole process takes a little more than a day.
The advantage of folding DNA into lumps is fully manifested here. All the numerous copies of the original ring inside each lump are identical. This means that the method is not very sensitive to errors in the manufacture of luminous molecules – even if there is an admixture of "unsuitable" molecules, its glow will be drowned in the light of dozens of "correct" indicators. For a single-molecular technology like the one used by Helios, such an impurity would mean an inevitable error, so the indicators must be manufactured with very high quality. In the case of nanosheets, quality is not so important, and this also reduces the cost of reagents.
In total, Complete Genomics specialists counted three complete human genomes. The first of them belongs to a white man of European origin, the second to a woman from Nigeria. Both of these genomes have been partially read as part of the HapMap project, and the female one should also soon be read completely as part of the 1000 Genomes project. The third DNA was extracted from a blood cell-a precursor of a patient with blood cancer. The scientists chose these samples to compare their results with known sequences and genetic anomalies. The purity of the method based on these data is estimated by Complete Genomics at less than one error per 100 thousand nucleotides.
For people, not for horses
It is worth noting that the new technique works only if there is a so–called reference sequence for the genome under study - for example, a DNA sequence defined for another representative of the same or very close biological species. Although DNA fragments are read multiple times (from 40 to 80 times), they are the same fragments all the time, and also very short.
It is impossible to put together a library of the complete genetic code from these short pieces of 60-70 letters. So in order to determine the genome of new species (for example, the horse described in the same issue of Science), you have to use other methods in which, if you use the same analogy, the library is first copied repeatedly, and then torn into shreds of several lines.
However, for medical and research applications, Complete Genomics technology is just what you need. Firstly, because the reference human genome has been around for six years. Secondly, because from a medical point of view, the differences between genomes from each other are the most interesting. And finally, because $4,400 is already quite a reasonable price available to many buyers.
However, the company does not work with private buyers – it is going to take orders only from large offices that already offer their clients services for determining and interpreting their DNA. Now it costs about $100 thousand. Let's see how this amount will change in the coming years.
Portal "Eternal youth" http://vechnayamolodost.ru09.11.2009