Sequencing Resources
1
Plasmidomics: Cloning Naturally Occurring Plasmids for Sequencing and Functional Analysis
Anne Marie Erler[1], Patricia Sobecky[2], Gary L. Andersen[1], and Peter Agron[1] (argon1@llnl.gov)
[1]Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory, Livermore, CA; and [2]School of Biology, Georgia Institute of Technology, Atlanta, GA
Plasmids are nonessential, stably maintained, extra-chromosomal DNA molecules commonly found in microbes. They imbue a great variety of traits on their hosts including resistance to toxic metals, increased ultraviolet radiation resistance, use of alternative food sources such as organic waste from human activity, virulence and antibiotic resistance. Therefore, nature has “pre-packaged” genes encoding traits of great interest, and the ability to take advantage of this cloning work would be of great value. A simple approach has been developed that potentially allows any circular plasmid to be established in Escherichia coli, thus facilitating sequencing and functional analysis. In vitro transposition was used to introduce a selectable marker as well as a plasmid replicon that is functional in E. coli. This way, circular plasmids that recombine with the donor molecule will replicate in the heterologous host when introduced by transformation regardless of the natural ability to do so. Three transposon donors, each tailored for different experimental approaches, were constucted. A small archaeal plasmid was established in E. coli, and testing with the recombinants provides strong evidence that this plasmid cannot normally replicate in the bacterial host. In addition, an approximately 60-kb uncharacterized plasmid from a South Carolina salt marsh bacterium was established in E. coli and was subsequently shown to have the natural ability to replicate in this host. Copy number amplification in E. coli easily produced quantities of this plasmid for direct DNA sequencing, which revealed high similarity to restriction-modification systems, suggesting one putative function for this plasmid in the environment.
2
Shotgun Library Utilization for Sequencing Projects at the JGI
Chris Detter (detter2@llnl.gov), Eileen Dalin, Jamie M. Jett, Doug Smith, Jenna Morgan, Hope Tice, Saima Shams, Corey Chinn, Eddy Rubin, and Paul M. Richardson
U.S. Department of Energy Joint Genome Institute, 2800 Mitchell Drive, Walnut Creek, California 94598 USA
The U.S. Department of Energy Joint Genome Institute has produced shotgun libraries for sequencing projects from sources including cosmids, fosmids, BACs and whole genomes including numerous microbes. Other large genomes that have been subcloned for sequencing include Fugu rubripes, Ciona intestinalis, Phanerochaete chrysosporium, Thalassiosira pseudonana, Populus trichocarpa, and Xenopus tropicalis. Standard subclones for these projects have been produced by random shearing of DNA followed by size selection (typically 3 kb inserts) and ligation into pUC18. Most of these libraries produce adequate assemblies with few uncaptured gaps. However, some problematic projects require more specialized libraries to span gaps and produce longer-range contiguity. A major need in the Production Sequencing process, especially for larger whole genome shotgun projects, is the use of medium insert libraries that span repeat sequences. Having paired end information from clones that span repeats greatly improves sequence contiguity in whole genome assemblies. For these reasons, our R&D group focused on developing a robust and reproducible 8-10 kb library construction protocol that could be scaled for use in our high-throughput sequencing facility.
Here, we will describe the construction, sequencing, and analysis of medium insert (8-10 kb) libraries used for sequencing at the Joint Genome Institute. The libraries were constructed from randomly sheared BAC or whole genomic DNA that was size selected and cloned into a kanamycin resistant low-copy plasmid (pCUGI21blu). To date, more than 100 BACs, as well as several microbial and larger whole genomes, have been successfully subcloned using these methods. Individual subclones produced by this method can be sequenced without modification of the high-throughput production process. Analysis of initial sequencing results indicates the inserts are stable and have a narrow distribution around the expected size. These libraries have been useful for assembling BAC clones spanning highly repetitive human sequence. Specific protocols and results will be described.
This work was performed under the auspices of the U.S. Department of Energy, Office of Biological and Environmental Research, by the University of California, under Contracts No. W-7405-Eng-48, No. DE-AC03-76SF00098, and No. W-7405-ENG-36.
3
BAC Libraries for Whole-Genome Sequencing, Comparative Genomics and Haplotype Analysis
Pieter J. de Jong (pdejong@mail.cho.org), Baoli Zhu, Mikhail Nefedov, Chung-Li Shu, Yuko Yoshinaga, and Kazutoyo Osoegawa
Children’s Hospital and Research Center Oakland (CHRCO)
Bacterial Artificial Chromosome Libraries are enjoying a continuing diversification of applications in sequencing, in functional genomics and diagnostics. BACs are used as the start for clone-based shotgun sequencing. BAC-ends provide scaffolding information for whole-genome sequence assembly. BACs also permit prioritized sequencing of chromosomal regions or gene families, hence being at the starting point for comparative genomics. Increasingly, BACs are used for gap filling in difficult regions, for haplotype analysis and for functional complementation. We will present an overview of the current strategies for BAC cloning and the growing number of species represented by BACs. About 70 BAC libraries have been prepared in our laboratory for an equivalent number of species and strains. To resolve difficult genomic regions, BACs are cloned with presumed minimal bias from sheared genomic fragments, resulting in “sheared” BACs from Drosophila melanogaster, mouse (C57BL/6J) and human DNA.
Work performed under the auspices of various funding sources, including USDA, NIH, NSF and DOE.
4
Using Transposons to Resolve Repeats in BAC Clones
Jamie Jett[1] (jmjett@lbl.gov), Jeremy Schmutz[2], Eileen Dalin[1], Jane Grimwood[2], Corey Chinn[1], Doug Smith[1], Susan Lucas[1], Chris Detter[1], Paul Richardson[1], and Eddy Rubin[1]
[1]US Department of Energy Joint Genome Institute, 2800 Mitchell Drive, Walnut Creek, California 94598 USA; [2]The Stanford Human Genome Center; and Department of Genetics, Stanford University School of Medicine, Stanford, CA 94305 USA
At the U.S. Department of Energy Joint Genome Institute, standard 3 kb shotgun libraries from tiling path BAC clones were created and sequenced for human Chromosomes 5, 16, and 19. For the most part, assemblies were complete with few uncaptured gaps. However, there were a number of individual BAC clones that stood out as problematic and unfinished due to highly repetitive regions composed of both tandem repeats and duplications within the clone. Chromosome 16 stood out as being especially repetitive.
Medium-size insert (8-10 kb) libraries were created to span problematic regions in an effort to capture unique sequence at both ends to help close gaps and flank repeat regions. Several hundred specifically selected BAC clones were sheared, size selected to 8-10 kb, subcloned and sequenced to a 10x depth. Following this process, many BAC clones were assembled to provide contiguous sequence. There were, however, a small number of important clones with unresolved assembly problems caused by direct tandem and duplication repeats that exceed the medium-subclone size. These problematic regions required an alternative method to verify large repeats (>16 kb) and identify possible unique sequence in the middle of each 8 kb subclone.
Using paired end sequence, ~500 medium-sized insert clones were selected from a total of 21 individual BAC clones for a transposon-based DNA sequencing strategy. Transposons, artificial transposable elements that integrate into the interior of DNA, allow for directed sequencing within a targeted subclone. The insertion of a transposon within internal regions of the medium-size clone provided sequencing priming sites throughout the targeted sequence for initiating a set of bi-directional di-deoxy ladders.
Here, we will describe DNA isolation, transposon insertion, sequencing, and analysis of bi-directional sequencing of internal regions within the 8-10 kb plasmids. To date, the paired bi-directional sequence information has assisted in the finishing of 9 BAC clones with 12 more in process.
This work was performed under the auspices of the U.S. Department of Energy, Office of Biological and Environmental Research, by the University of California, under Contracts No. W-7405-Eng-48, No. DE-AC03-76SF00098, and No. W-7405-ENG-36.
5
Single Molecule DNA Sequence Profiling In Zero-Mode Waveguides Using g-Phosphate Linked Nucleotide Analogs
Jonas Korlach (jk109@cornell.edu), Michael Levene, Mathieu Foquet, Stephen W. Turner, Harold G. Craighead, and Watt W. Webb
Cornell University, School of Applied Physics, Clark Hall, Ithaca, New York 14853
We show that nucleotides with a fluorescent label attached to the g-phgosphate constitute a superior class of compounds for successful observation of single molecule DNA polymerase activity in zero-mode waveguides. In contrast to base-labeled analogs (Levene et al., Science 299: 682), the fluorophore is cleaved from an incorporated base upon polymerization, providing a low background necessary for detection of successive enzymatic turnovers. The enzyme is not inhibited, permitting replacement of all four bases with spectrally separable analogs.
Using these analogs, single molecule DNA synthesis by the highly processive DNA polymerase from phage f29 was measured in zero-mode waveguides. These nanostructures drastically reduce the observation volume to tens of zeptoliters, thereby enabling single fluorophore detection at high concentrations (>1µM) where polymerase-mediated DNA synthesis is fast and processive. Incorporation is detected by longer residence times of analogs in the active site (~1-10ms), compared to diffusion (<100µs). One-base sequence patterns on synthetic sequences were obtained. Strategies for efficient excitation of spectrally separable analogs involving FRET pair coupled dyes for de novo single DNA molecule sequencing are presented.
DoE DE-FG02-99ER62809 & NCRR-NIH P41-RR04224.
6
Complete Direct Sequencing of BAC, Phage and Microbial Genomes using ThermoFidelase, Fimer and D-Strap Technologies
S. Kozyavkin (serg@fidelitysystems.com), O. Shcherbinina, V. Shakhova, N. Pavlova, A. Morocho, V. Karamychev, Y. Malykh, A. Pavlov, N. Polouchine, A. Malykh, and A. Slesarev
Fidelity Systems, Inc., 7961 Cessna Avenue, Gaithersburg, MD 20879-4117 http://www.fidelitysystems.com
We have developed tools and workflow for direct genomic DNA sequencing that eliminates the need in subcloning and production of shotgun libraries, minimizes the number of sequencing reactions and dramatically accelerates the assembly of complete sequence. Using our approach we have completed sequencing of many BAC, phage and microbial genomes.
A core component of the procedure is the use of genomic DNA as a template in a robust sequencing reaction. The addition of ThermoFidelase 2 with its unique combination of topoisomerase and DNA binding activities is used to shorten the cycles of denaturation and primer annealing. The dramatic increase in specificity, quality and yield of priming from megatemplates is achieved by using Fimers (modified oligonucleotides with proprietary SUC modifications) instead of regular primers and multiplying the number of thermal cycles. The third element of new strategy, D-Strap is based on Fimer design that targets evolutionary conserved elements. The advantages of direct genomic sequencing include elimination of cloning artefacts and library or PCR cross contamination that are extremely important for production sequencing of closely related organisms, non-biased complete and low coverage of the genome that results in significant savings on data processing. In addition, the use of D-Strap Fimers in multiple projects contributes to cost savings and has a potential for the fastest way of complete sequencing of closely related species.
We have completed five types of BAC projects starting from full BAC shotgun, skimming shotgun, whole mouse genome shotgun, cDNA sequence and more recently, composite (reference) human genome. In all cases robust sequencing reactions on BAC DNA using ThermoFidelase and plates of Fimers were used for finishing. The projects were somewhat different in workflow organization. The new requirements for individual’s genome sequencing (NO ERRORS) are much stricter than current (Bermuda) requirements for composite human genome project (1 error per 10 kb). The no errors goal was achieved by complete direct sequening of BAC provided by Dr. Larionov (NCI). The assembly of Fimer directed reads was straightforward, completely automated and took less than a minute. The results indicated that TAR method of BAC production and direct complete BAC sequencing using ThermoFidelase and plates of Fimers can be scaled up for NO ERRORS sequencing of genomes of many individuals for a fraction of cost and effort compared to that of shotgun or PCR based methods.
Shotgun method of sequencing 200 kb phage genomes is unproductive because of the under representation of significant portion of phage sequences in shotgun libraries and the up front cost of library production. In contrast, Fimer directed sequencing of phage DNA turned out to be a straightforward method fully compatible with numerous modifications in phage genomic DNA. We have completed NO ERRORS sequencing of a number of phages. In addition, we have designed and validated a set of D-Strap Fimers that target weakly conserved phage genes. The developed set of D-Strap Fimers is useful for initiation of phage genome sequencing and multi locus sequence typing of phages.
A high fraction of unfinished microbial genomes and long time required for their finishing clearly illustrates deficiencies of whole genome shotgun method. ThermoFidelase and Fimer based direct microbial DNA sequencing was used by us and other teams to accelerate finishing of shotgun projects. Our recent goals have been concentrated on optimization and scale up of direct microbial sequencing. Last year we have completed and published Whole Genome Direct sequencing of M. kandleri AV19. Using D-Strap method we have sequenced another isolate of this species. Our recent work on finishing Lacto genomes provides us with an opportunity to compare deep shotgun vs direct genome sequencing mehtods. Our experience indicates that deep shotgun and increased complexity of the shotgun libraries does not automatically give a complete genome sequence. Many gaps, misassemblies, low quality regions and uncertain repeat structure are all present. At the same time the data processing load on assembly, contamination clean up and chimera detection increases exponentially.
We expect that minimizing shotgun coverage and implementation of direct genome sequencing into production setting is a viable alternative to the current practice of deep shotgun drafting of microbial genomes that can produce complete sequences without escalating project costs. In collaboration with JGI we have demonstrated that direct sequencing of microbial genomes using ThermoFidelase and plates of Fimers is compatible with 96 capillary sequencers. The two remaining items needed for complete direct sequencing of microbial genomes that does not rely on shotgun clones are the ability to sequence genomic DNA samples of poor quality and resolution of sequence repeats that are longer than read length. We have found the solution to both problems by developing a new method of direct genomic sequencing based on capture of Sanger fragments. To increase the yield of capture we have developed novel biotin amidites with super-long linkers that meet additional requirements on minimal charge density and optimized rigidity and flexibility. We will present the examples of using biotinilated Fimers with super-long linkers for elimination of background fluorescent noise and long repeat resolution in direct microbial genome sequencing.
Supported in part by DOE (DE-FG02-98ER82577, 00ER83009).
7
Isolation of Exceptional Chromosomal Regions to Close the Gaps in the Draft Human Genome Sequence
S.-H. Leem, N. Kouprina, and V. Larionov (larionov@mail.nih.gov)
Laboratory of Biosystems and Cancer; National Cancer Institute, NIH, Bethesda, MD 20892
The Human Genome Project has entered a final phase during which the sequence must be completed, corrected and finalized. During this phase of the project, sequence gaps must be closed and the overall quality of the sequence improved. To carry this out, it will be necessary to collect additional sequence data. It is not clear whether all of the sequences missing from the draft human genome sequence are represented in the bacterial libraries. Recent works on sequencing of Plasmodium and Dictyostelium genomes has shown that a high AT content of the chromosomes prevents the construction of large insert BAC libraries. The gaps between contigs in the draft of human genome may also arise from chromosomal regions that are not present in the Escherichia coli libraries used for DNA sequencing because they can not be cloned efficiently, if at all, in bacteria. To address this question, ten gap regions from human chromosomes 5, 16 and 19 were recovered in yeast as circular YAC/BACs with a selective TAR cloning method. Further analysis of the gap isolates revealed two types of sequences: a) those that were unstable both in YAC and BAC forms, b) those that were stable in yeast but toxic for bacterial cells. Sequencing of these exceptional regions required non-standard approaches. Some clones were analyzed using a BAC direct sequencing strategy. Other clones were sequenced in YAC forms to avoid transfer of YAC clones to bacterial cells, where they undergo deletions and rearrangements. This work helped to close last four gaps on chromosome 19. Closing the gaps on chromosome 16 is in progress. In summary, this work and other reports indicate that alternative cloning systems and hosts may be critical to complete the final phase of the Human Genome Project. One of such approaches, TAR cloning in yeast, will allow for rapid and selective isolation of targeted regions of the human genome that can not be verified or completed using clones generated and propagated in E. coli.
8
pFOS-LA: A Modified Vector for Production of Random Shear Fosmid Libraries
J. Longmire[1] (longmire@telomere.lanl.gov ), N. Brown[1], S. Malfatti[2], Jack Meeks[3], and Patrick Chain[2]
[1]Bioscience Division, Los Alamos National Laboratory, Los Alamos, NM; [2]Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory, Livermore, CA; and [3]Section of Microbiology, University of California, Davis, CA
The standard fosmid cloning vector pFOS-1 was modified to allow simple and rapid cloning of 40 kb fragments generated by random shearing. The modified vector is termed pFOS-LA. A double-stranded oligonucleotide was ligated into the HindIII site of pFOS-1 to provide two, unique blunt end sites (PmlI and SwaI). Cloning in the modified vector proceeds as follows: Target DNA is homogeneously sheared to 40 kb, end repaired (to make blunt ends), dephosphorylated (to prevent chimeric inserts) and ligated into either the SwaI or PmlI site of the modified vector. Ligation products are packaged, infected and plated using methods developed for pFOS-1 (Kim et al., [1992] Nucleic Acid Research 20: 1083-1085). The modified vector retains advantages afforded by pFOS-1 including single copy origin of replication, double cos site design, high cloning efficiency and ability to make partial digest libraries if so desired. Distinct advantages of the modified vector include 1) A more random cloning approach resulting in libraries with potentially better sequence representation; and 2) Starting DNA can be smaller in size since it does not have to be partially digested prior to cloning. The relaxed requirement for very high molecular weight target DNA allows cloning of DNA samples that are already sheared to 30-40 kb upon isolation. Such DNA would be difficult or impossible to clone by partial digestion.
We have used pFOS-LA to construct a random shear library for the 9.5 Mb genome of the microbe Nostoc punctiforme. Cloning 0.2 ug of N. punctiforme genomic DNA yielded a library containing 95,000 independent clones with inserts averaging 41.3kb (by fingerprint analysis) and provides >400-fold coverage. As will be discussed, a portion of this library is currently being used to finish sequence the genome of N. punctiforme.
It is anticipated that the modified vector will be especially advantageous for mapping and sequencing genomes that are difficult to clone by partial digestion, such as the genomes of many bacterial species where restriction sites are not evenly distributed or where these sites are blocked by methylation. Given the high cloning efficiency and unbiased cloning strategy, this vector should be useful for making deep coverage, high representation “metagenome” libraries from complex microbial communities.
This work was performed under the auspices of the U. S. Department of Energy by the University of California, Los Alamos National Laboratory.
9
Using YACs to Close Gaps at the JGI
Jenna Morgan (jlmorgan@lbl.gov), Duncan Scott, Joel Martin, Tijana Glavina, Susan Lucas, Chris Detter, Paul Richardson, and Eddy Rubin
U.S. Department of Energy Joint Genome Institute, 2800 Mitchell Drive, Walnut Creek, California 94598 USA.
Yeast Artificial Chromosomes (YACs) have been used for over a decade to construct representative large insert DNA libraries for several genome sequencing efforts (e.g. human, mouse, rice, Arabidopsis). YACs confer several advantages over alternative methods for constructing physical maps of large genomes. The YAC vector allows the ligation of insert DNA fragments of over 1,000 kilobases. YAC clones are propagated in yeast as stable artificial chromosomes, and can maintain sequences that are unstable in prokaryotic-based cloning systems. In some cases (e.g. Dictyostelium discoideum), YACs may represent the only viable method for the construction of large insert libraries.
YACs present unique challenges in practice, however. The larger inserts tend to lose integrity during transmission in yeast, suffering from deletions and chromosomal rearrangements. Although frequency of occurrence varies among libraries, YACs often contain chimeric inserts. Additionally, YACs are present in single copies and are very similar in size to endogenous yeast chromosomes, thus complicating YAC DNA isolation. In part for these reasons, the JGI has relied almost exclusively on more “user friendly” cosmid and BAC libraries for sequencing its chromosomes 5, 16, and 19. In general, this approach has been extremely successful; however, as the Human Genome Project draws to a close, there are a few regions of the chromosomes that are not spanned by sequenced BAC clones. These regions, referred to as “clone gaps” or “Type 3” gaps, require alternative approaches for closure.
Here we present our strategy for using YACs to close these Type 3 gaps in human chromosomes 5, 16, and 19. The challenge to increase low YAC DNA yield was tackled by scaling up standard YAC isolation protocols. The optimization of pulsed-field gel conditions to separate YACs ranging in size from 150kb to 1700kb was the most onerous task. To date we have successfully closed several Type 3 gaps in each of our human chromosomes using YAC subclone libraries.
Recent advances in YAC cloning technology have greatly improved the efficiency of the YAC shotgun sequencing effort. The transformation-associated recombination (TAR) cloning strategy developed by Larionov et al. has produced gap-spanning YAC clones without chimerism and without the need for the construction of an entire library of random clones. Recently we produced a subclone library that was used to close the last remaining clone gap on chromosome 19 using a circular YAC clone. Currently, work is in progress to produce circular YAC clones for each of the remaining clone gaps in chromosomes 5 and 16.
This work was performed under the auspices of the U.S. Department of Energy, Office of Biological and Environmental Research, by the University of California, under Contracts No. W-7405-Eng-48, No. DE-AC03-76SF00098, and No. W-7405-ENG-36.
10
Helix-Hairpin-Helix Motifs to Create Processive, Hyperstable and Inhibitor-Resistant Enzymes
Alexei Slesarev (alex@fidelitysystems.com ), Andrey Pavlov, Nadya Pavlova, and Sergei Kozyavkin
Fidelity Systems, Inc.
We have developed a technology for creating DNA metabolizing enzymes with the enhanced processivity, outstanding stability, remarkable tolerance to high salt concentrations and resistance to chemical and biological inhibitors. We demonstrate the potential of this technology on DNA polymerases. Our method consists in creating chimeras composed of core polymerase domains or entire unmodified enzymes fused with different helix-hairpin-helix (HhH) domains derived from topoisomerase V (TopoV) of Methanopyrus kandleri (1,2). HhH is a widespread motif involved in sequence-nonspecific DNA binding. There are 24 such motifs in Topo V (3), most of them being dispensable for the activity of TopoV, yet their removal greatly affects the stability and salt concentration range of TopoV. We demonstrate that different HhH cassettes fused with either NH2-terminus or COOH-terminus of the Stoffel fragment of Taq polymerase or with Pfu polymerase increase the resistance of the DNA polymerase activity to high salt concentrations, up to 0.5 M NaCl or 1.8 M KGlu. The processivity of chimeric polymerases increases and depends on the structure of HhH attached to the catalytic domains. Anions play a major role in the inhibition of DNA polymerases. Chimeras are more thermostable than their unmodified counterparts and show no loss of the activity after incubation at 100°C for at least 1 hour. Moreover, the chimeras are able to extend primers at least at 105°C. Our approach to raise the salt tolerance of polymerases and their stability also allows for cycle sequencing and PCR at high salt concentrations and at temperatures inaccessible for other DNA polymerases.
This work was supported by grants from DOE and NIH.
1. Andrey R. Pavlov, Galina I. Belova, Sergei A. Kozyavkin, and Alexei I. Slesarev. PNAS 2002 99: 13510-13515
2. Slesarev, et al. PNAS 2002 99: 4644-4649
3. Galina I. Belova, Rajendra Prasad, Sergei A. Kozyavkin, James A. Lake, Samuel H. Wilson, and Alexei I. Slesarev. PNAS 2001 98: 6015-6020
11
Efficient Isothermal Amplification of Single DNA Molecules
Stanley Tabor (tabor@hms.harvard.edu) and Charles Richardson
Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02115
We are developing DNA polymerases for use in DNA sequencing and amplification applications. We will describe our progress in developing a very efficient isothermal amplification system that is based on the replication machinery of bacteriophage T7. This system is capable of amplifying single DNA molecules, increasing the amount of DNA more than a trillion-fold in a 30 min reaction. Amplification is nonspecific. The template can be circular (e.g. plasmid or BAC DNA) or linear (e.g. genomic DNA). The products are linear double-stranded DNA fragments that average several thousand base pairs in length. The reaction requires the T7 DNA polymerase, the T7 helicase/primase complex (T7 gene 4 protein), and single-stranded DNA binding protein. The reaction requires no exogenous primers, using the inherent priming activity of the T7 primase. It is critical to remove all contaminating DNA from the reaction mixture, since all DNA present will be amplified. We have developed a successful strategy to cleanse the reaction mixture of contaminating DNA by pretreatment with Micrococcal nuclease.
We will present our progress on the use of this system in the following applications:
1. The preparation of plasmid and BAC DNA templates for DNA sequencing.
2. The whole genome amplification of rare DNAs, such as hard-to-culture microorganisms and purified chromosomes.
3. An extremely sensitive and rapid assay to determine the total amount of DNA in a sample, with a linear range of over 13 orders of magnitude.
4. The sequencing of haplotypes by the amplification of DNA from single chromosomes.
12
Improved DENS: Finishing Without Custom Primers—From Human to Microbes
Olga Chertkov, Marie-Claude Krawczyk, Mira Dimitrijevic-Bussod, David Bruce, Mark Mundt, Paul Gilna, Norman Doggett, and Levy Ulanovsky (levy@anl.gov)
Los Alamos National Laboratory
DENS (Differential Extension with Nucleotide Subsets) is primer walk sequencing without custom primer synthesis. DENS largely eliminates the cost of custom primer synthesis—several dollars, compared to less than a dollar for the rest of the expenses (per lane) combined. DENS works by converting a short primer (selected from a pre-synthesized library of 1440 octamers with 2 degenerate bases each) into a longer one on the template at the intended site only. DENS starts with a limited initial extension of the octamer primer at 20° C in the presence of only 2 of the 4 possible dNTPs. The primer is extended by 5 bases or longer at the intended priming site, which is deliberately selected, as is the two-dNTP set, to maximize the extension length. The subsequent cycle-sequencing at 60° C accepts the primer extended at the intended site, but not at alternative sites where the initial extension (if any) is generally short. We have now automated all labor-intensive steps in DENS and have employed this as part of our finishing strategy to improve low quality targets. Our current throughput is 16,000–20,000 DENS reactions per month. Much of human chromosome 16 and several bacterial genomes have been finished using > 140,000 DENS reactions with the success rate rising from ~40% to ~80%. DENS accounts for more than half of finishing at LANL.