Talk:Single-molecule real-time sequencing
![]() | This article has not yet been rated on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||
|
Copyright
The lead may be plagiar of [1]. Synchronism (talk) 06:04, 7 November 2008 (UTC)
Application
The language here, implying the technology will be applicable to a variety of biological problems, reads like PacBio marketing literature. In particular, AFAIK there is no scientifically documented basis for the statement "The longer read length allows de novo genome sequencing and easier genome assemblies."
- well, it may be pacbio marketing lit, but re "The longer...." I don't have time to find it right now, but pretty sure there is a lot of science to back that up - might find it in critiques of helicos, which has short read lengths. —Preceding unsigned comment added by Cinnamon colbert (talk • contribs) 14:01, 3 November 2009 (UTC)
I would suggest that this entire section be removed until there are published examples of these applications being accomplished with the technology (which many, myself included, would welcome).Rwintle (talk) 18:05, 2 April 2009 (UTC)
- Hm. So, 1k holes and 1.5k bp total read length at ~10 bp/sec. So within three minutes and 1.5 Mb later, the device becomes useless? Mycoplasma genitalium (Venter's latest sequencing project) has roughly 580 kbp, but that's the world record for the smallest genome yet discovered. Maybe my math is wrong, or the numbers hate me, but I am inclined to agree that whole genome sequencing (via this method) might not be an accurate claim for anything other than the world's smallest genomes. Although, I'm sure you could claim that if you used a million holes, you could increase the amount of DNA sequenced, sure. I haven't looked over the references yet. -- kanzure (talk) 21:53, 27 May 2009 (UTC)
- Their current web page claims 75k holes per cell - Owlmonkey (talk) 07:09, 3 December 2010 (UTC)
Talks at American Society of Human Genetics 2010 in Washington, D.C. suggest SMRT sequencing currently provides ~80% read accuracy. Not reasonably high enough to justify claim that PacBio is able to sequence genomes in a de novo manner. —Preceding unsigned comment added by 12.43.245.254 (talk) 06:07, 6 November 2010 (UTC)
- They claim 99.99% consensus reads on this web page and specifically discuss de novo, perhaps their lower raw read accuracy is easier to correct then other sequencing techniques with typical repetitive coverage, leading to reasonable consensus accuracy. - Owlmonkey (talk) 07:02, 3 December 2010 (UTC)
spam
this whole thing is an advert for pacbio, and needs to be rewritten, and mainly, merged into the exisitng DNA sequencing articleCinnamon colbert (talk) 12:51, 15 October 2009 (UTC)
- Now that the technology is becoming popular for bacterial and fungal assembly because of its unique qualities, and used for researching copy number variation in cancer research for similar reasons, seems reasonable to me that it has a unique place in the history of sequencing. But if you see some language that is not neutral please improve it. Owlmonkey (talk) 14:38, 25 September 2013 (UTC)
Lots of potential for improvement here
The article has potential but needs some work. No, I mean lots of work. Regarding the lead section, it could become shorter and more understandable to non-experts. Maybe I can help, but my sources are written in Russian. Can we use non-English sources for the English Wikipedia? Χρυσάνθη Λυκούση (talk) 01:23, 26 March 2014 (UTC)
- Non-English sources are better than no sources at all, of course, but only when English sources aren't available (see WP:NOENG). I agree that some more layman-friendly text is in order. Any new sources would be welcome! §everal⇒|Times 13:03, 26 March 2014 (UTC)
Sequencing Performance
There is a lot of info in the Sequencing Performance section about how read length exceeds that of Sanger sequencing, but what about read accuracy? Sanger is regarded as the gold standard for accurate sequencing. How does the PacBio approach compare? My own understanding is that the error rate is relatively high, at about 12% for 1x coverage (see, for example http://cshl.edu/Research/PacBio.html).