CHI's second next-generation sequencing conference showcases the first personal genomes and future technologies.
Dec. 17, 2007 | PROVIDENCE, RI - It's been 20 years since scientist-entrepreneur Kevin Ulmer launched the first single-molecule DNA sequencing company, SEQ. Since then, Ulmer has studied or consulted for just about every next-generation sequencing system. Now he has been born again. The title of his keynote address at CHI's Exploring Next Generation Sequencing conference*, said it all: "Sanger sequencing is dead: Long live Sanger sequencing!"
After two decades, Ulmer is returning to the personal genomics field with a new venture called Genome Corp. (see sidebar: "Genome Corp. Born Again"). Though the data in GenBank is doubling every 16 months, and the cost halves approximately every 25 months. Ulmer believes another dramatic shift in technology would be required to make the "$1000 genome" a reality before 2040.
Ulmer offered his the pros and cons of all four leading next-generation sequencing technologies: pyrophosphate diffusion and high consumable costs (454); read-length limits and image processing (Illumina, Applied Biosystems); and photobleaching, monochromatic signal detection, and cost (Helicos). (On that final point, Ulmer quoted a $2 million estimate published by the Wall Street Journal, whereas Helicos subsequently priced its HeliScope at $1.35 million.)
Ulmer has concluded that the proven track record, unparalleled read lengths, and well understood error properties of traditional Sanger sequencing serve as fundamental assets in building the equivalent of a "printing plant" for genome sequencing. Without divulging details, Ulmer said his variation on Sanger will involve single-molecule processing, "one pot" in vitro amplification, and a rethink on electrophoresis and imaging.
The meeting was notable for presentations on the first two complete human genome sequences. David Wheeler (Baylor College of Medicine) discussed the first draft of James Watson's genome, based on more than 100 million fragment reads performed by 454. Samuel Levy (J. Craig Venter Institute) recapped the publication of the "first human diploid reference sequence," that of Venter himself, using Sanger methods. Both speakers emphasized the vast amount of genetic variation in these sequences, not merely SNPs but insertions/deletions and copy number variants.
Meanwhile, several start-up companies presented new platforms that hope to challenge 454, AB, Illumina, and Helicos in the coming years. One was Reveo, one of the first groups to enter the $10 million Archon X PRIZE for genomics. James Eakin, manager of business development, claimed that Reveo's nano-knife edge technology has the potential to sequence a human genome "error-free, in minutes, for pennies." He presented "a disruptive instrument concept" that uses physical, non-disruptive methods, rather than chemical, to directly interrogate chromosomal DNA sequences.
Reveo's solid-state instrument relies on concepts borrowed from the semiconductor, optics, and micro-fabrication industries. An array of electro-conductive, nano-knife edge probes, each about 10 nm long and less than 1 nm thick, interrogates the DNA, including epigenomic modifications. Eakin envisions a microwave-sized instrument costing about $5,000, with standard microprocessors reading sequence data at a speed of 1 base/nanosecond.
Steven Gordon, CEO of Intelligent Bio-Systems (IBS), is readying a platform based on sequencing-by- synthesis chemistry (SBS) developed by Jingyue Ju's lab at Columbia University. The target market is clinical "diagnostic sequencing," with a premium placed on "cost per test" rather than "cost per base." IBS' "pinpoint" system uses single-stranded DNA templates on an array and incorporates fluorescently labeled nucleotide analogs into complementary strands. The chemistry involves custom cleavable reversible terminators and dyes. After signal measurement, both terminator and dye are cleaved before the cycle repeats.
Gordon says his firm's bench-top instrument will produce about 5 GB sequence per day, and is looking for early access customers. The instrument will sell for about $275,000, enabling many labs beyond the genome centers to tackle whole genome analyses.
Another promising contender is NABsys (Nanopore array-based systems), a spin-out of Brown University. VP John Oliver explained how his start-up is developing hybridization assisted nanopore sequencing, combining the strengths of sequencing-by-hybridization (SBH), which "works but doesn't scale," with nanopore sequencing, which "scales but doesn't work." While others insist that nanopore sequencing could yield DNA sequences in real time, Oliver is content to use nanopores at a lower level of resolution - to detect the location of hybridization probes to the target sequence.
The key is to scale up the method by building nanopore arrays to permit rapid determination of positional hybridization information. The assembly of the target sequence relies on algorithms similar to those used in SBH.
Perhaps the most entertaining presentation came from Leonard Bloksberg, founder of New Zealand-based Cartesian Gridspeed (see "The Quest to Make Sequence Sense," Bio-IT World, November 2006). "Biologists are feeling the pain of the massive data overload required to even participate in modern biology," said Bloksberg. His flagship program, dubbed SLIM Search, works up to four orders of magnitude faster than BLAST with high volumes of short sequences, and has affordable academic seat licenses. With BLAST laboring away in the background on his laptop, Bloksberg illustrated the acceleration of sequence analysis using his proprietary program.
* CHI's next Next-Generation Sequencing event switches to the West Coast and takes place April 23-24, 2008, in San Diego.
Subscribe to Bio-IT World magazine.