The Scientist is linking to an imaginary1 article from PNAS in which researchers compare the cost of sequencing microbial (I'm guessing they mean bacteria) genomes using the traditional Sanger method and the hot new technology developed by 454. Not so surprisingly, they find that a hybrid method -- ~5x coverage with Sanger followed by a couple rounds of 454 -- is the most cost effective strategy. I say it's not surprising because usually some intermediate solution wins out in science.
The Sanger method is nice because you get paired end reads from clones, which are very helpful in assembling shotgun sequenced genomes. If you thought shotgun sequencing was messy (compared to sequencing mapped clones), then strategies that take advantage of the 454 technology will make Pig-Pen look like Sean Connery to you. Researchers are currently using 454 to resequence bacterial genomes, and I think some work has been done using 454 for de novo genome sequencing in bacteria. The technology is currently not tractable for de novo sequencing of eukaryotic genomes, but this study suggest that 454 may prove useful in a clean-up role after some shotgun Sanger sequence has been generated.
1- They claim the article exists, and it just hasn't been posted on the PNAS webpage yet.
- Log in to post comments
I think 454 is being used to sequence neandertal autosomes as well. I'm guessing only for certain genes of particular evolutionary interest
To be fair, 454's sequencing technology was only published on last year, and it's still a very new technology. Sanger has been around for decades and has probably reached its maximum potential. I think as 454 improves the machines and refines the method, it will probably one day replace the Sanger method for genomic sequencing. It's just going to take some time to increase their read lengths.
Also, what does this say about the speed of sequencing? I thought the neatest aspect of the 454 when it was published last summer in Nature was its incredible speed in determining the genome they sequenced, not necessarily cost.
Here is the article you want:
http://www.pnas.org/cgi/content/abstract/0604351103v1
They claim a 20% reduction in cost, mostly by reducing the number of sanger reads necessary.