and Duane Bartley
The Johns Hopkins University
This three-and-a-half day meeting in Hilton Head, SC on September 16-20, 1995 included discussion on sequencing strategies, newly engineered polymerases to carry out these strategies, and automating as much of the effort as is "humanly" or "mechanically" possible. Researchers presented summary reports describing the H. influenzae Rd (1) and Mycoplasma genitalium (2) sequencing efforts and updates outlining the progress of the E. coli, yeast, B. subtilis, Drosophila, C. elegans, Arabidopsis, mouse, and human genome projects. Genome research efforts are moving towards refining sequencing technologies and gaining an increased understanding of genome organization and function from analysis of sequence data ("in silico" experimentation).
Any strategy chosen to complete a large-scale sequencing project has pros and cons. Some groups strongly favor a directed over a shotgun sequencing strategy, while others strongly support the opposite approach. A shotgun strategy typically follows the completion of a high resolution physical map and allows rapid accumulation of data with high redundancy. A primer-walking strategy does not require multiple clonings or template preparations and produces minimally redundant sequence information. The primary disadvantages of primer walking include the cost of synthesizing new primers to extend the sequence and the delay in synthesizing these new primers. These disadvantages are being overcome by researchers using "primer libraries". Strategies for high-volume sequencing using a primer library composed of short oligonucleotides sufficient to prime multiple DNA sequencing reactions are being developed. Overall, there is more intermingling of the two sequencing strategies, allowing more rapid development of contiguous sequences and then gap closure.
Two new thermostable DNA polymerases--Thermo-Sequenase (developed by Amersham) and AmpliTaq FS (developed by Roche Molecular Systems and Perkin Elmer)--were introduced and compared in sessions on sequencing chemistry and technology. Carl Fuller and colleagues from Amersham described ThermoSequenase, a mutant of Taq DNA polymerase developed by Stanley Tabor and Charles Richardson at Harvard (3) with a single amino acid substitution that increases the reactivity of ddNTPs to approximately that observed for dNTPs. Sandra Spurgeon and colleagues from Perkin Elmer identified the same amino acid substitution using genetic screening in the development of AmpliTaq FS. Both genetically altered enzymes incorporate dye terminators as efficiently as dNTPs, requiring less dye-terminators than presently used for cycle sequencing. Additionally, unincorporated dye terminators can be removed via a precipitation step, in place of a spin-column. Both enzymes produce significantly more even band intensities, relative to other thermostable DNA polymerases, and this property should contribute to more accurate base calling and heterozygote detection.
In fluorescent sequencing with four dyes, each dye has a different electrophoretic mobility that can obscure sequence information in the raw data file. A computer algorithm that corrects for these different gel mobilities extracts sequence information from the raw data file. Unfortunately, DNA fragments migrate differently in different size classes, and the algorithms were originally written to interpret data within a narrow size range. Thus, the current dye chemistry and computer algorithms define sequence read length limits. However, if each dye migrated similarly, each DNA fragment would migrate similarly, and one should be able to extract more sequence information from the raw data files. This is why Richard Gibbs and colleagues substituted "BODIPY" dyes (Molecular Probes) for ABI's FAM, JOE, TAMRA, and ROX in fluorescently labeled primers. The advantages of the new dyes are: i, each alters the mobility of the DNA in a similar manner, eliminating the need to compensate for mobility differences; ii, each emits a similar level of fluorescence, so they can be used quantitatively; and iii, each has a more narrow emission spectrum than the conventional dyes, so they are more easily resolved.
Much progress in automation, detection, and measurement was reported, resulting from collaborations between biologists and engineers. Eric Lander and Trevor Hawkins showed videos of robotic systems to facilitate high-throughput PCR mapping and sequencing. Glen Evans and Harold Garner are generating fluorescent sequence data directly from cosmid DNA prepared via "Dr. Prepper", a DNA preparation robot developed in their laboratory. David Lockhart and Robert Lipshutz described high-density arrays of oligonucleotide probes that can be used to determine individual genetic variation and for light-directed oligonucleotide synthesis. Futuristic approaches involving direct sequencing of a single DNA molecule by degradation and spectrophotometry of the released nucleotide were described by Kevin Ulmer. Mark Burns and colleagues are designing a thermal-cycler and sequencer on a computer chip using micro-machine technology. Elizabeth Stewart described the use of a panel of genome radiation hybrids to construct a high resolution map of human genes. Andre Marziali described an automated, high-throughput method for preparing sequencing templates. Deidre Meldrum demonstrated the accurate measurement of 90 picoliter volumes. The abstracts for the meeting can be found in Genome Science & Technology, Volume 1, Number 1, 1995.
References
1. Fleischmann, R.D., et al. (1995) Science 269, 496-512. "Whole-Genome Random Sequencing and Assembly of Haemophilus influenzae Rd."
2. Fraser, C.M., et al. (1995) Science 270, 397-403. "The Minimal Gene Complement of Mycoplasma genitalium."
3. Tabor, S. and Richardson, C.C. (1995) PNAS 92, 6339-6343. "A single residue in DNA polymerases of the Escherichia coli DNA polymerase I family is critical for distinguishing between deoxy- and dideoxynucleotides".
Return to the The ABRF Home Page