A new method that combines DNA sample preparation from hair shafts with whole-genome shotgun sequencing on 454’s platform may find applications in population genetics, ancient DNA studies, and forensics, according to experts.
But they caution that the approach needs to be further refined in order to be widely applicable in these fields.
Last week, scientists at Pennsylvania State University and the University of Copenhagen led a team that published the new method in Science. In the paper, they described the use of the approach to sequence entire mitochondrial genomes from 10 Siberian mammoth specimens with up to 48-fold coverage.
What made this high coverage possible is the fact that mitochondrial DNA seems to be better preserved in hair than chromosomal, or nuclear, DNA, according to Stephan Schuster, senior author of the study and an associate professor of biochemistry and molecular biology at Penn State.
Nevertheless, the study “sets the stage for a nuclear genome project of mammoth,” Schuster said, suggesting that nuclear DNA can be sequenced from hair in a similar fashion.
The DNA in the hair shaft seemed to be well-protected both from degradation and from contamination with microbes because the scientists were able to extract sufficient amounts of DNA even from a mammoth hair sample that had been stored at room temperature for 200 years.
Schuster sees three main applications for the new method: population genetics of current and extinct species; analysis of ancient DNA, for example from museum collections; and forensics.
Entire museum collections are open to large-scale DNA sequencing now, Schuster believes. “You could go back and look at the genetic diversity of all the Darwin finches for which feathers are stored in museums,” he said. The work would not do a lot of damage to the samples, because unlike extracting DNA from bone, using hair or feathers does not destroy any anatomical information.
To be sure, sequencing DNA extracted from hair shafts is nothing new for museum scientists. “We have been doing that for 15 years,” according to Robert Fleischer, who runs the center for Conservation and Evolutionary Genetics at the National Zoological Park of the Smithsonian Institution in Washington, DC.
Fleischer and his colleagues have been using standard PCR coupled with Sanger sequencing to amplify short regions of mitochondrial and nuclear DNA from sub-fossil bones and hair or feathers from museum specimens, he told In Sequence this week.
If whole-genome shotgun sequencing could provide data for, say, 50 different genes to compare, instead of just one, that would be an advantage for both population genetics and phylogenetic studies, he said.
However, unless the sequencing coverage is very high, scientists cannot be sure that shotgun sequencing yields data for all these genes in every sample. “You have to do a lot of runs in order to guarantee you are going to get the same genes from every individual,” Fleischer said.
Nevertheless, if scientists could develop methods to target large numbers of specific DNA regions of interest with 454 sequencing, the method would become widely applicable, he believes. “Then the sky is the limit,” said Fleischer. “That would make it a lot cheaper and a lot more useful for the things we do.
“I think it has great utility in the future, with some tweaking,” he added. “But at the present time, it doesn’t seem like it’s that useful for a phylogenetic study or a population genetic study, unless it’s [only] mitochondrial DNA [that] you are interested in.”
The picture changes, though, if a genome of a closely related species already exists. For example, any shotgun sequence data from an extinct elephant species could be compared to the African elephant’s genome, which he said will soon be completed.
At the moment, Fleischer’s center owns a ABI 3100 sequencer, which it has had for three years. “Hopefully, we are going to get an upgrade to a bigger, better model next year,” he said. A 454 sequencer, with its $500,000 price tag, is not an option at the moment, though.
Would his center have use for a high-throughput sequencer? “Yes, we would,” Fleischer said. “Not just for the ancient material but more for the modern material” like malaria samples, which his center is also analyzing.
For forensic applications, the lower cost of 454’s technology, as well as the simplified sample prep, make it “a wonderful advancement” over Sanger sequencing, according to Mitchell Holland, an associate professor of biochemistry and molecular biology and associate director of the forensic science program at Penn State.
“If you could routinely sequence the whole mitochondrial genome in a very cost-effective way ... for forensics samples, then that would have great benefit because it’s a little bit more discriminating; and if it’s at a lower cost, that means you could do it more readily.”
At the moment, a sequence-based forensic analysis of an older bone or hair sample costs between $1,500 and $3,000 per sample, he said. “If you could pool enough samples together, [454 sequencing] could be an overall cost-effective way to go.”
But the method would still need to be validated and streamlined for routine work in a forensics lab, Holland said. “If you work at the research bench, a lot of the time what you do is very different than what you do when you get into an operations environment, meaning generating standard protocols that have been validated and that work every time, and that you can apply that to forensic casework, the results of which end up in court.”
The question also is whether entire mitochondrial genomes are necessary for the purpose of distinguishing one individual from another.
“The challenge with the 454 technology is to balance the brute force of the technology and so much sequencing going on in one run … with the need to generate only a certain amount of data that will be informative for forensics,” Holland said.
Right now, when forensic scientists analyze samples that do not contain a lot of chromosomal DNA, like hair shafts or fingernail clippings, they sequence a short mitochondrial genome region of 600 to 1,000 base pairs “that has the majority of the discrimination content that you can use to differentiate between individuals,” Holland said.
But there might be reason to sequence entire mitochondrial genomes instead. For example, two suspects might share an identical sequence in the short mitochondrial region. “Going outside of that region to the whole [mitochondrial] genome, you may find sites that differentiate between the two,” he said.
However, it is still unclear whether the 454 technology can work with the tiny amounts of DNA from forensic samples. “The problem in forensic casework is [that] you usually just have a single hair [fiber] that you find at a crime scene. Or even if you find three to four hairs, you don’t know if they came from the same person, so you don’t necessarily pool them together,” according to Holland.
In the mammoth study, the smallest amount of hair the scientists analyzed was 0.2 grams. However, according to Schuster, that is probably not the lower limit of the technology. “We believe we can go even lower,” he said.
Holland does not know a forensic DNA laboratory in the US – of which there are more than 500 – that uses next-generation sequencing at the moment.
But “if you could routinely sequence the whole mitochondrial genome in a very cost-effective way ... for forensics samples, then that would have great benefit because it’s a little bit more discriminating; and if it’s at a lower cost, that means you could do it more readily,” he said.