Funded with a $5.3 million grant from the National Science Foundation, a team of researchers led by the University of Delaware is planning to use Illumina’s next-generation sequencing technology to study the epigenetics of rice.
“We are hoping to develop a resource for understanding epigenetics in the cereal crops, which include plants like corn, wheat, and rice,” said Blake Meyers, an associate professor of plant and soil sciences at the University of Delaware and the principal investigator on the project.
Under the four-year grant, the researchers plan to use Illumina’s Genome Analyzer to study — on a genome-wide scale — DNA methylation, histone methylation, small RNA profiles, and mRNA profiles in several tissues of at least 10 different lines of rice.
One advantage of studying rice is its relatively small genome — 400 megabases — and the fact that the genomes of two different rice varieties have already been sequenced, Meyers said.
About $1 million of the $5.3 million award has been budgeted for sequencing, he said.
Meyers’ lab has already been working for several years with Illumina and its predecessors, Solexa and Lynx Therapeutics, on mRNA and small RNA analysis, using Lynx’s massively parallel signature sequencing technology. In the last few years, he has also used 454’s platform to study small RNA.
However, as of late, “the entire field [of small RNA profiling in plants] has shifted pretty dramatically to [Illumina’s] SBS [sequencing-by-synthesis],” he said. “It’s really a perfect fit for small RNA sequencing [in plants].”
Because plants, unlike animals, have large and complex pools of small RNA, “it’s very important to be able to sequence very deeply,” he explained, and “the depth with the Illumina technology is much greater than we could get from 454.”
The reads from Illumina’s sequencing technology — currently 35 base pairs — are also long enough to cover the entire length of microRNAs and small interfering RNAs, he said, which are generally less than 25 nucleotides long.
“We get more reads at a cheaper price [than with 454’s technology], while still getting the full-length sequence of the small RNA from each read,” he said.
A number of other research groups are involved in the project. Steven Jacobsen and Matteo Pellegrini from the University of California, Los Angeles, will lead the DNA methylation and histone modification studies. Jacobsen has already studied epigenetics in Arabidopsis, and “we are hoping to take advantage of his expertise in that area, and move some of the exciting discoveries he has made [in Arabidopsis] into rice biology,” Meyers said. “One goal is to get single-base resolution for DNA methylation.”
The UCLA scientists will also study histone modifications using chromatin immunoprecipitation coupled with sequencing, or ChIP seq, an approach that several other research groups have applied to studying human chromatin (see In Sequence 6/12/2007).
Meyers will focus on generating and analyzing small RNA and gene-expression profiles. He and his Delaware colleague Pamela Green collaborated with Lynx, later Solexa, for several years on sequencing small RNA using MPSS and published a paper on the application in Science in 2005.
The researchers also plan to develop a number of bioinformatics methods to analyze the sequence data. Pellegrini is focusing on tools to analyze DNA methylation and histone modification data. “We developed our own software to map the sequence data to the genome,” he told In Sequence last week. “We use part of the Solexa pipeline, but we have modified it considerably.”
Meyers’ lab, on the other hand, wants to improve databases and tools that his lab has already developed to analyze small RNA and mRNA data from another project, including a genome viewer. “[We will develop our viewer] to connect these very disparate data types, mRNA, small RNA, methylation, and histone,” he said.
Last year, Meyers and his colleague Green won an NSF grant to sequence small RNAs and microRNAs in 32 plants and algae (see RNAi News, In Sequence’s sister publication, 10/5/2006). Dealing with large amounts of data has been a challenge, he said. It took several months “to get all of our trimming and genome mapping and storage scripts and informatics up to speed.”
He and his team were used to dealing with large datasets from MPSS, he said, but not with many such datasets in a short time. Because Illumina sequencing costs less than MPSS, “we could now afford to do a lot more sequencing than we had done with MPSS,“ he said. As a result, “suddenly we were getting in new datasets every week, and with MPSS, it had been every few months. We had been able to take our time with MPSS; we can no longer afford that time.”
It is still unclear where the sequencing data for the rice epigenetics project will be generated. According to Meyers, the researchers might use Illumina’s sequencing service, but the University of Delaware might also acquire a Genome Analyzer, or his collaborators at UCLA might use their own instrument.
“For now, we are planning to use Illumina, but … if we learn of another method that is clearly superior, we would certainly weigh our options.”
Getting a machine in house “would certainly lower the cost of the reactions, which would allow us to do a little bit more science,” Meyers said.
On the other hand, he and his colleagues have been happy with Illumina’s sequencing services. “It would be hard for us to achieve a better quality, at least in the first six months of running our own machine,” he said. Also, since Illumina has several instruments available, “they don’t really have bottlenecks. If you have one machine, you are limited by the number of runs that you can do per year.”
According to Meyers, who has used Illumina’s sequencing service for other small RNA projects, the company charges him $10,000 for an entire run of small RNA sequencing and between $3,500 and $4,000 for a single channel of small RNA sequencing, depending on whether the scientists prepare the library. Illumina declined to comment on the accuracy of this pricing information.
Illumina has said in the past that reagents for a sequencing run cost approximately $3,000.
So far, the researchers have started to obtain rice lines and to develop rice mutants, and are testing the platform on wild type rice. Since rice has a generation time of four to six months, the first year of the project will be mostly devoted to creating the rice lines. Two other collaborators will be involved in that: Guo-Liang Wang, a rice biologist from Ohio State University; and Yulin Jia, a plant biologist at the Department of Agriculture’s Dale Bumpers National Rice Research Center in Stuttgart, Ark., who has experience in growing rice populations.
In the second and third year, the scientists are planning to acquire “a lot of the sequence data” that they plan to analyze in the third and fourth year.
While Meyers is currently planning to use Illumina’s sequencing platform for the entire project, he does not rule out switching to another technology in the future.
He has “talked to” ABI about its SOLiD platform but “until we actually have readily available data from SOLiD, I think people are going to focus on Illumina’s [technology] because it’s available.”
But Meyers said he would welcome competition from another platform. “Hopefully, that would drive the companies to improve their products even more, and some might offer advantages that others don’t have, whether that’s longer reads, or different types of paired-end reads, or different depths, or different cost structures,” he said.
“For now, we are planning to use Illumina, but we're more committed to doing the highest-quality science,” he told In Sequence by e-mail this week. “If we learn of another method that is clearly superior, we would certainly weigh our options.”