The Rise of Long Reads: Mendelspod Podcast Series
Thursday, September 11, 2014
Mendelspod host Theral Timpson kicked off a new podcast series this week on long-read sequencing that will include interviews with luminaries in the genomics field. Check out this introductory article from Timpson for an explanation of why scientists are demanding longer reads to meet their research goals.
The first interview is with Mike Snyder at Stanford, who has published recent papers in Nature Biotechnology and PNAS using Single Molecule, Real-Time (SMRT®) Sequencing for transcriptome analysis and demonstrated that long reads enable full coverage of RNA molecules. He discusses that work and his views on long-read sequencing and transcriptomics on the show. Here are some highlights:
On the state of transcriptomics
Without using long-read sequencing, the way transcriptomes are figured out is “crazy,” Snyder explains. “We take RNA, we blow it up into little fragments, and then we try and assemble them back together to see what the transcriptome looked like in the first place. And that’s a horrible way to do this because what we’re really trying to do is understand all of the different isoforms of a transcript….So when you blow them up and try to reassemble them back together you can’t always figure out which parts of the puzzle belong together.” (This reminds us of a clever cartoon in Nature Methods last year, subscription required.)
The power of long-read sequencing
By nature, long-read sequencing can avoid ‘blowing up’ the transcripts, because as Snyder has demonstrated in his studies, it is possible to generate full-length transcripts using SMRT Sequencing.
“The power of long-read sequencing is really to be able to capture all of the information in its intact form without trying to solve a jigsaw puzzle that you may have put together wrong.” Snyder explains that misassembling transcripts can make it impossible to understand what is going on. For example, different isoforms of the same tumor gene have very different functions and may be either healthy or oncogenic.
On the $1000 genome
“Has the race to the $1000 genome been at the expense of quality?” Timpson asked Snyder. “Yes,” he replied. “I think people’s eyes are opening to that.” Based on what is currently considered the “$1,000 Genome” (which Snyder points out cannot actually be commercially purchased today for $1,000), he says, “the quality is still not there, there’s still significant gaps.”
“People don’t realize this but there are still several hundred gaps in the human genome that have never been closed,” Snyder explains. And, ironically, in the process of attempting to fix these gaps in the reference genome, researchers end up uncovering more errors, “so the number of gaps in the human genome has stayed fairly constant over the last 10 years or so,” he adds.
Snyder believes that PacBio’s SMRT Sequencing provides one solution to this problem, by spanning gaps, resolving structural variation and providing the ‘gold standard’ in quality in sequencing today.