Some DNA just won’t reveal its sequence. Known as “difficult templates,” these molecules resist scientists’ efforts to decipher their bases — at least using standard protocols. But the Association of Biomolecular Resource Facilities’ DNA Sequencing Research Group is now taking on these problem DNAs in a new study.
The aim of the project, which follows on similar studies the DSRG conducted in 2003, 2000, and 1998, is to develop guidelines for sequencing various kinds of difficult templates. Study participants are asked to try their own protocols first, then apply the most successful approaches in a second phase. The study results will be presented at the ABRF 2008 annual meeting in Salt Lake City in February.
Jan Kieleczawa, who is organizing this year’s study, has an easy definition for difficult templates: “If you apply standard ABI sequencing conditions, and you cannot sequence it, it’s difficult,” he said. “You have to apply some kind of modification [to the protocol] in order to get through this difficult region.”
But there are no easy solutions, and no tried-and-true protocols yet. “I know that there is no one-size-fits-all protocol,” said Kieleczawa, a lab manager and principal scientist at Wyeth in Cambridge, Mass. Rather, researchers have to come up with individual approaches for each type of tricky sequence.
Difficult templates are not an unusual challenge: at least 10 percent of the samples his sequencing facility sees are difficult, he said.
“There are templates out there that really challenge our abilities, some more so than others,” said Kevin Knudtson, director of the DNA facility at the University of Iowa. “I think we are all looking for solutions how to handle these various difficult templates.”
In order to develop such solutions, the DSRG selected eight templates, each containing at least one difficult-to-sequence region, to be analyzed by Sanger sequencing.
Among them are a moderately GC-rich template, with a GC content below 75 percent, and a very GC-rich template, with more than 95 percent GC content in a stretch of about 100 bases.
Other templates contain a homopolymer, an Alu-repeat, a hairpin, and long stretches of two bases that are not repeats.
The group recently sent these templates, along with sequencing primers, to 30 interested sequencing labs in the US, Canada, and Australia for analysis. Template availability and logistics limited the number of groups that could participate, which were chosen on a first-come-first-served basis.
In the first phase of the study, the labs, which did not know what the templates were or what difficulties they contained, were asked to apply any protocol of their choice in triplicate to sequence them.
Kieleczawa and his colleagues are currently analyzing the results from the first phase. Besides determining the Q20 values of the data, the DSRG also asked about experimental parameters, such as instrumentation, cycling conditions, purification methods, and formulations.
“Maybe people have come up with ways to address these difficult templates, and maybe this is going to be an opportunity to share these.”
Based on the results, the DSRG is going to select the most successful protocols — up to three for each category of difficult template — and will ask the same participants to sequence the templates again with these protocols to see if they get similar results.
The group plans to present the study results at the upcoming ABRF meeting in Salt Lake City in February, and plans to make the protocols widely available later on.
Sequencing conditions will probably be most important, according to Kieleczawa. “What really matters is chemistry,” he said. “Instruments should not matter, because almost everybody, if not 100 percent, use ABI’s [sequencers].”
Past DSRG studies have also tackled the problem of difficult templates, most recently in 2003, when the group focused on templates containing extensive repeat sequences. But no study has addressed such a variety of difficulties before.
Also, “the sequencing chemistries have changed quite a bit,” Knudtson said, whose lab does not participate in the study. Over the years, “maybe people have come up with ways to address these difficult templates, and maybe this is going to be an opportunity to share these,” he said. “Maybe there is now a solution [available] that’s a little bit more consistent in dealing with some of these various difficult templates.”
Knudtson hopes that his lab will benefit from the results: “This could be a point of reference, a place to start on how to deal with some of these difficult template types.”
But not all are optimistic. Harold Hills, director of DNA sequencing and the enzyme freezer program at the University of Massachusetts Medical School in Worcester, and a study participant, said he already has a protocol that works well for GC-rich templates, and he is not sure the study is going to benefit his lab.
Being a small laboratory, he said the study is costing his group “a small fortune,” and some of the samples yielded no sequence data.
Others, however, believe studies like this are valuable. Difficult templates may only represent a “minor percentage overall, but for the individual wanting data from a difficult template, it is a big problem,” said another East Coast-based core facility director and study participant who asked to remain anonymous because she was not authorized to speak on behalf of her institution.
“We will look at the results to see if any novel solutions are included,” the facility director told In Sequence via e-mail. “It is also good to have a way to compare our results to others.”