Researchers at the National Heart, Lung, and Blood Institute have used Solexa sequencing to profile the distribution of methylated histones and other DNA-binding proteins across the human genome.
The researchers published their results in the May 17 issue of Cell, making this the first peer-reviewed article involving Illumina’s sequencing technology to appear in print.
The article represents a milestone of sorts for Illumina: Several researchers have said in the past that the only way to judge next-generation sequencing technologies fairly is by relying on peer-reviewed data.
The NHLBI researchers, led by Keji Zhao, a senior investigator in the laboratory of molecular immunology, profiled the distribution of 20 different histone lysine and arginine methylations, histone variant H2A.Z, RNA polymerase II, and insulator binding protein CTCF in human white blood cells.
According to Zhao, the “major discovery” for histone methylations was that most of the monomethylations correlated with gene activation instead of repression.
Using Illumina’s Genome Analyzer, which the lab installed in mid-January, they sequenced DNA from chromatin immunoprecipitation, or ChIP, experiments. The lab ran the instrument non-stop, performing two runs per week for about 10 weeks, according to Zhao. By mid-April, “we got all the data for this paper,” he said.
In the past, Zhao’s lab, which is interested in how chromatin modifications control gene expression or other functions of the genome, had combined ChIP with serial analysis of gene expression, or SAGE, a method that relies on traditional Sanger sequencing.
However, that method, also known as genome-wide mapping technique, or GMAT, suffers from “relatively low resolution of 500-1,000 bp and the considerable cost of sequencing in order to cover the entire genome,” according to the Cell paper.
According to Zhao, it cost $80,000 and took about three months to analyze a single histone methylation by traditional sequencing, which generated about one million tags.
Using Illumina’s platform, he said, one modification costs between $1,000 and $1,500, and takes less than one week to generate between 5 million and 10 million tags. The researchers can analyze up to three different modifications in a single run on the Genome Analyzer, Zhao said, which creates between 20 million and 30 million tags.
“Without this rapid and economic sequencing strategy we cannot really analyze this many different modifications in such a short time,” Zhao told In Sequence last week. “And without analyzing so many, we cannot really draw [such] a strong conclusion.”
The instrument’s current read length of up to 36 base pairs is more than enough for analyzing ChIP DNA, according to Zhao. “We never needed to go that long; we just need 25 base pairs,” he said. “What we need is only to find a tag that can be identified in the genome.” About 20 percent of the tags, he said, cannot be mapped uniquely because of repetitive regions in the human genome.
“Without this rapid and economic sequencing strategy we cannot really analyze this many different modifications in such a short time.”
Zhao developed an early interest in next-generation sequencing because his research “requires this huge amount of sequencing,” and he was aware of the technology since before Solexa — which Illumina acquired earlier this year — merged with Lynx Therapeutics in 2005.
He said his lab chose Illumina’s instrument because it “provided the most convincing data.” In fact, other researchers have found that Illumina’s technology is well suited to analyze ChIP DNA, including three groups that presented their research at the recent Biology of Genomes meeting at Cold Spring Harbor Laboratory (see In Sequence 5/15/2007).
Martin Hirst, a researcher at the British Columbia Cancer Agency Genome Sciences Center in Vancouver who presented at the meeting, told In Sequence via e-mail this week that the NHLBI study “would not have been possible without the use of second-generation sequencing” because the cost would have been “prohibitive” without the Illumina platform.
Hirst added that his group used random shearing to generate DNA fragments prior to immunoprecipitation, which is a departure from the NHLBI team, which used MNase treatment.
“It remains to be seen what difference, if any, this will make to data interpretation,” he said.
Funding for the NHLBI instrument came from intramural resources, after an outside review panel recommended that the institute acquire next-generation sequencing technology.