June 4, 2002
Berkeley Lab Science Beat Berkeley Lab Science Beat
Hybrid Magnet Boosts Genomic Sequencing Rates
Berkeley Lab Science Beat

Lab website index

Lawrence Berkeley National Lab home page

Search Lab science articles archive
 Advanced Search  
Search Tips

To meet the demands of today's high-volume genome sequencing projects, Berkeley Lab researchers have developed a powerful magnet that obtains DNA samples much faster than commercially available magnetic plates.

The magnet, which borrows from more than a decade of research that informed the development of particle accelerator magnets, can be applied to several emerging biotechnology fields. In addition to gene sequencing, it can be used in functional genomics, which is the study of what genes do. And it can be used in proteomics, which is the study of how proteins both activate and function in organisms.

It's based on a hybrid approach in which permanent magnets are coupled to ferromagnetic materials. This combination produces magnetic fields that are much stronger than those produced by magnet plates based solely on permanent magnets, according to David Humphries of Berkeley Lab's Engineering Division.

David Humphries with the hybrid magnet he helped develop for biotechnology applications

So far, Humphries and colleagues have used the magnet to more efficiently conduct several sequencing initiatives at the Department of Energy's Joint Genome Institute (JGI). In the lab, it's used in a common process that draws genetic material out of solution. The technique uses hand-sized plastic plates, called microtiter plates, that are pitted with 384 tiny wells. Each well is filled with a solution that contains DNA. Next, tiny magnetic particles measuring one micron (one millionth of a meter) in diameter are added to the solution. These beads are coated with a carboxylate layer that is specially chosen because it binds with DNA. In this manner, the carboxylate-coated bead attaches to the DNA, and the DNA is magnetized. Finally, the microtiter plate is placed on the magnet plate, and the DNA is drawn to the bottom of each well and out of the solution.

Until recently, this process relied on relatively unsophisticated commercial magnets that are adequate for small benchtop applications, but are hard-pressed to accommodate high-volume processes in which millions of base pairs of genes are sequenced. Such volumes are increasingly required in comparative sequencing efforts in which the genomes of relatively simple organisms are compared with the largely finished human genome. For example, the JGI recently sequenced the sea squirt, which at one stage in its life cycle possesses the most primitive notochord, or vertebral column, of any animal. Understanding how this structure develops in sea squirts may shed light on how it develops in humans.

To undertake these mind-bogglingly massive comparative genome studies, engineers have developed sophisticated liquid handling robots that deposit DNA-containing solutions into microtiter plates, and high-speed capillary electrophoresis sequencing machines that read the DNA once its pulled from the solution. However, the middle step -- drawing the DNA out of solution -- often looms as the weak link because of slow, weak magnets.

That's where Berkeley lab's hybrid magnet comes in. It boasts a field strength, or magnetic pull, that at its surface is 70 percent stronger than the most powerful commercial magnet. And one centimeter above the magnet, the magnet's pull is 400 percent stronger than commercial magnets used with 384-well plates, and 1,000 percent stronger than commercial magnets used with 96-well plates. This means the magnet quickly and thoroughly grabs DNA floating near the top of even thick, viscous solutions. And it securely holds DNA at the bottom of wells, especially important given that liquid handling robots subject solutions to considerable turbulence.

The hybrid magnet, foreground, attracts strands of DNA attached to magnetic particles in a microtiter plate, background.

In addition, the magnet's gradient, or the change of its field strength over distance, can be manipulated. This means magnetized particles can be held at user-defined positions in the microtiter wells, enabling researchers to selectively separate different particles from a solution.

Together, the magnet's improved field strength and gradient distribution translates to unprecedented production rates. Industrial sequencing processes are judged by a quantitative yardstick called read length, which is the number of base pairs of genes that are sequenced in a single cycle. Of equal importance is its pass rate, a qualitative assessment that measures the amount of acceptable data obtained in the sequencing process. The higher this percentage, the less data must be discarded and the more efficient the technique. Today, reading 500 high-quality base pairs per cycle is considered good. The JGI production sequencing process, using the hybrid magnet, produces read lengths of 630 base pairs with pass rates above 90 percent.

"That's an industry milestone," Humphries says. "No facility we know of comes close to these read lengths and this sample rate."

The revved-up magnets have helped double the JGI's sequencing capacity. The lab's previous sample processing system employed 96-well microtiter plates and pure permanent magnets. This highly automated system produced roughly 20 million sequenced bases per day, a level that was one of the highest of any sequencing laboratory.

The current sample processing system employs both 384-well microtiter plates and the new hybrid magnet technology. Today, JGI researchers sequence more than 40 million bases per day. In addition, the system's improved efficiency has enabled the lab to reduce both the process's space and personnel requirements.

Look for these sequencing rates to get even faster. The next step, according to Humphries, is to tailor the hybrid magnets to fit next-generation microtiter plates with 1,536 wells.

"The motivation behind these magnets is to sequence as fast as possible," Humphries says. "And although the initial goal of the human genome project is largely accomplished, there is a tremendous amount of work left to be done."

Berkeley Lab's Martin Pollard and Livermore Lab's Chris Elkin, along with a number of JGI and Berkeley Lab employees, greatly contributed to the technology's success.