NEW YORK, Feb. 13 (GenomeWeb News) - 454 Life Sciences' sequencer may show great potential for de novo sequencing of microbial genomes, but it can't replace Sanger sequencing completely, according to researchers from the Department of Energy's Joint Genome Institute.
In fact, they said that 454's Genome Sequencer 20 works well with Sanger sequencing for this kind of research to produce a high-quality sequence. The researchers presented their findings on this hybrid approach at the Advances in Genome Biology and Technology Conference, held last week at Marco Island, Fla.
"454 can generate high quality draft assemblies, not final finished sequences," Feng Chen, group leader of technology development at the Lawrence Berkeley National Laboratory, told GenomeWeb News. "The quality of [the] 454 assembly is reasonably high, but it is still lower than the required quality standard for [a] finished sequence."
The JGI researchers said the Genome Sequencer 20 cannot replace a Sanger-based machine for microbial genomes because of limitations caused by short read length and a lack of paired-end sequencing information. The paired-end information from Sanger sequencing is turning out to be crucial in scaffolding and gap closure, the researchers said.
"The 454 system currently is not designed for paired-end sequencing," said Chen. "It sequences single-stranded DNA fragments, and when the run is done, DNA is disposed. 454 is working on another approach to get paired-end information."
The fact that 454's technology doesn't require any cloning is one of the reasons the technology is so fast, but there are advantages to having clones. "454 does not provide any clones covering the gap, so one [might] not have materials to work on for gap closure," explained Chen. "The other part of the finishing is to polish the low quality areas; again, one would need clones covering these areas to work on."
Because 454 sequencing cannot fully replace Sanger sequencing, JGI researchers decided to test how the two could be used together most efficiently. Using the microbial genome Prochlorococcus marinus NATTL2A, they combined 454 sequencing results with different amounts of paired Sanger sequencing reads from three different sized shotgun libraries. They then analyzed the results.
The scientists discovered that combining 12X to 14X coverage of 454 sequencing with 3X to 5X coverage of Sanger sequencing can produce a high-quality sequence. One four-hour run and three days of work produces the 14X coverage from the Genome Sequencer 20. By comparison, it takes about a month to produce this kind of coverage using Sanger sequencing.
The researchers point out that while the 454 run was supposed to provide 20X coverage, the coverage turned out to be 14X when only the reads aligned with reference sequencer were counted. Coverage loss was due to the low quality of some of the reads that never made it to the assembly stage.
Some reads also went to small contigs with low assembly quality. The researchers say that going above 14X coverage with 454, when using the hybrid strategy, did not improve the quality of the assembly.
"454 has great potential and we are going to use 454 for most of the microbial genomes we are going to sequence," Chen said. "JGI will use (the) hybrid strategy, but will not try to use 454 alone to attempt to get finished sequences."
Kate O'Rourke covers the next-generation genome-sequencing market for GenomeWeb News. E-mail her at [email protected].