Created: 1st December 1999, last updated: 21st February 2000, © 1999 ABRF
M. Roxane Bonnera and Linda Wood Ballardb
aLaboratory of Molecular Systematics and Evolution, University of Arizona, Tucson, AZ; bGenomics Core Facility, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT
Because analysis of single nucleotide polymorphisms (SNPs) can be invaluable in understanding genomic variation and the genetic basis of disease, there is a need for high-throughput, high-accuracy mutation detection methods for identifying SNPs. A sequencing core facility can enhance the services it offers by providing genome analysis methods to search for informative SNPs. Denaturing high-performance liquid chromatography and single-strand conformation polymorphism analysis are methods of mutation detection that are amenable to a sequencing core environment. They are useful for screening large sample sets to identify novel SNPs, eliminating the need to sequence every sample in the set. These methods allow analysis of more samples than would otherwise be economically feasible by sequencing alone. (J Biomol Tech 1999;10:177-186)
Key Words: mutation detection, genomic variation, heteroduplex, conformation polymorphism, denaturing high-performance liquid chromatography (DHPLC), single-strand conformation polymorphism (SCCP) analysis.
Address correspondence and reprint requests to M. Roxane Bonner, Arizona Research Labs, Division of Biotechnology, University of Arizona, P.O. Box 210088, Tucson, AZ 85721 (email: bonner@u.arizona.edu).
DNA sequencing technology has advanced greatly as a result of the many genome sequencing projects underway. Sequencing core facilities have benefited from these advances and have been able to enhance their services by providing longer reads, quicker turnarounds, and reduced sequencing costs. Although completion of these genome projects can further the understanding of the genetic structure of each genome, it cannot address issues of genetic variation within and between the respective genomes. Knowledge of genetic variation is important in understanding evolutionary processes and in identifying the genetic basis of disease, and it has significant impact on genetic counseling, forensics, and environmental management.
Projects designed to study genomic variation usually involve large sample sets, but the cost of sequencing to gain this information in most cases is prohibitive. Such projects underlie the necessity for rapid and sensitive methods other than sequencing for identifying genomic variation. Sequencing core laboratories can take advantage of this need by offering alternative services for analyzing genomes that do not directly involve sequencing.
Many alternatives to sequencing involve comparative analysis of genomes. The polymerase chain reaction (PCR) is used to amplify fragments 100 to 1000 base pairs (bp) long, which are analyzed to detect insertions, deletions, and single nucleotide polymorphisms (SNPs). These comparative methods are supplemented by sequencing, because after a potential mutation or polymorphism has been identified, it can be confirmed and characterized only by direct sequencing. These methods can screen a large set of samples for potential mutants, and then only those samples identified as such need be sequenced.
Single-strand conformation polymorphism (SSCP) analysis and denaturing high-performance liquid chromatography (DHPLC) are two comparative methods that are easily adaptable to the sequencing core environment. SSCP identifies variation by distinguishing changes in the secondary structure of a single-stranded PCR product in response to changes in nucleotide sequence.1 SSCP can be run on automated sequencers and is therefore a good candidate method for adoption by an established core facility. DHPLC identifies variation by detecting heteroduplex formation between reannealed wild-type and mutant PCR fragments.2 It has the advantage of being fully automated, with high accuracy, high sensitivity, and high throughput, all of which are important in a large-volume core facility. The equipment used for DHPLC has the additional benefit of allowing analyses of oligonucleotide quality control and quantification. It can also be used for PCR quantification and cleanup before sequencing or cloning. We have successfully incorporated DHPLC and SSCP in the services of our molecular core facilities and are therefore able to offer an assessment of the advantages and disadvantages of each method.
DHPLC identifies mutations and polymorphisms based on detection of heteroduplex formation between mismatched nucleotides in double-stranded PCR amplified DNA. Sequence variation creates a mixed population of heteroduplexes and homoduplexes during reannealing of wild-type and mutant DNA (Fig. 1). When this mixed population is analyzed by HPLC under partially denaturing temperatures, the heteroduplexes elute from the column earlier than the homoduplexes because of their reduced melting temperature.3 Analysis can be performed on individual samples to determine heterozygosity (Fig. 2) or on mixed samples to identify sequence variation between individuals.4
FIGURE 1. Heteroduplex formation. Wild-type (Wt) and mutant (Mt) PCR products are heated to denature each strand and then allowed to cool slowly. The result is a mixed population of the original homoduplexes plus heteroduplexes containing the mismatched bases.
FIGURE 2. Identifying heterozygosity. The PCR product from a heterozygote naturally forms heteroduplexes because of the sequence variation of each allele. This heterozygosity is easily identified by DHPLC when analyzed under partially denaturing temperatures. Conditions: eluent A: 0.1 M triethylammonium acetate (TEAA); eluent B: 0.1 M TEAA, 25% acetonitrile; linear gradient 35% to 55% B in 1 minute, 55% to 65% B in 5 minutes, 100% B for 1 minute, 35% B for 2 minutes; 56°C; 260 nm; 0.75 mL/minute.
DHPLC is an ion-pair reversed-phase high-performance liquid chromatography (IP-RP-HPLC) method.5 There are two columns on the market for performing fragment analysis: the DNASep Column (Transgenomic, Inc., Omaha, NE, USA) and the Zorbax Eclipse dsDNA Analysis Column (Hewlett Packard, Palo Alto, CA, USA). The Eclipse column packing is composed of spherical silica particles (Zorbax Rx-SIL type B), densely layered with an aliphatic organosilane stationary phase. The DNASep column has a stationary phase composed of a polystyrene-divinylbenzene copolymer. The ion-pairing agent used with either column is triethylammonium acetate (TEAA), which mediates binding of DNA to the stationary phase. Acetonitrile is used as an organic agent to achieve subsequent separation of the DNA from the column.
DHPLC requires a biocompatible HPLC gradient pumping system with a flow path composed of PEEK tubing, fittings, and filters; an ultraviolet (UV) detector capable of 254 or 260 nm; and a column oven capable of maintaining temperatures of 50°C to 80°C. An autosampler configured to handle 96 0.2-mL microtubes is optimal for automated processing of samples.
Transgenomic, Inc., markets the WAVE Nucleic Acid Fragment Analysis System, a complete, fully automated system for performing fragment analysis. The system is completely computer controlled, offering all the specifications listed previously, including fragment analysis software. An optional fraction collector can be added to the system to automate the isolation and purification of fragments as they elute from the column. The WAVE system has been incorporated into the core facility at the University of Arizona.
Because DHPLC is performed on crude PCR product, no preanalysis cleanup step is required. However, if PCR is performed under oil, it is recommended that the oil be removed before analysis to extend the life of the column. Primer modifications are unnecessary for DHPLC, avoiding the need for expensive end-labeled oligos.
The sensitivity and accuracy of DHPLC have been reported to be 92% to 100% for PCR fragments 198 to 732 bp.6,7 We have investigated the accuracy of DHPLC for detecting known point mutations in 300- to 850-bp PCR fragments. Shown in Figure 3 is the analysis of a known SNP that was amplified in different sizes of PCR products and then analyzed by means of DHPLC to test whether the mutation was identified in these increasingly longer fragments. This C-->T transition was reliably detected in DNA fragments up to 651 bp. Similar tests on other known SNPs have shown that resolution of sequence variation in fragments larger than 500 bp is not always achieved and depends on the nature of the mutation. From these results, we have concluded that the optimal size range for detecting SNPs by DHPLC with 100% accuracy is 100 to 500 bp.
FIGURE 3. Assessing size limitations for DHPLC. This C-->T transition is easily resolved in fragments of 357 bp and 535 bp. Analysis of a 651-bp fragment shows subtle resolution of the mutation.
Resolution of heteroduplexes is temperature dependent and is directly related to the GC content of the fragment being analyzed. Optimization of DHPLC requires generating a melting profile for each PCR fragment to determine its optimal temperature for analysis. This can be done empirically by running a control sample at several different temperatures to find the temperature range for which the fragment shows partial denaturation, as indicated by a reduction in retention time when temperature is increased (Fig. 4). Normally, the lowest temperature that shows a change in retention time of -1 minute is optimal for identification of sequence variation. Alternatively, the optimal temperature can be mathematically calculated if the sequence of the fragment is known. The WaveMaker software by Transgenomic predicts the required acetonitrile gradient and optimal temperature for any imported DNA sequence. The optimal temperature can also be calculated by using the DHPLC Melt Program written by researchers at Stanford University (program can be accessed at http://insertion.stanford.edu/melt.html).
FIGURE 4. Optimizing denaturing high-performance liquid chromatography. Analysis of a control sample under increasing temperatures identifies the range in which this PCR product is partially denatured. The optimal temperature for resolving heteroduplexes is usually the lowest temperature at which denaturing begins to be evident, as indicated by a reduction in retention time. In this example, the optimal temperature for analyzing samples is 56°C.
Throughput can be increased by multiplexing fragments that fall into this size range.8 In multiplexing, the fragments must differ in size by more than 100 bp to allow adequate separation of potential heteroduplexes. They must also be of similar GC content, because fragments with widely varying percentages of GC require different acetonitrile concentrations and different temperatures to achieve SNP identification. In a properly designed experiment, multiplexing allows the analysis of about 1000 bp, rivaling the accuracy and throughput of automated sequencing, although at a fraction of the time and cost per sample.
The Laboratory of Molecular Systematics and Evolution (LMSE) at the University of Arizona charges $1.50 per sample, which covers chemicals, cost of the column, and annual maintenance agreement on the machine. This assumes 2000 injections per column, although the DNASep columns used in the LMSE have performed well after 4000 injections. An average run time per sample is 6 minutes for a single-fragment assay, and one fully automated DHPLC machine can analyze up to 240 samples in a 24-hour period. The average run time for a multiplex assay is 9 minutes, allowing the analysis of only 160 samples in 24 hours but increasing the number of base pairs analyzed by as much as 25%.
An IP-RP-HPLC system can be used to perform analyses other than DHPLC. Under fully denaturing temperatures (60°C to 70°C), single-stranded products can be separated to run quality control assays on oligonucleotides.9 Oligos can be evaluated for the presence of misincorporations that may have occurred during synthesis. Depending on the ion-pairing agent used, oligonucleotides can be separated strictly according to size (Fig. 5) (Haefele R, Gjerde D. Application note 103, 1999, Transgenomic, Inc.). This is useful for analyzing synthesized oligos for the presence of failure sequences or other impurities. The isolated fragments can be collected to produce purified oligos suitable for PCR or sequencing. This method also can be used to assess the integrity of a primer showing signs of degradation. The nondegraded fraction can be separated and collected, avoiding the need for resynthesis.
FIGURE 5. Oligonucleotide quality control. At denaturing temperatures, three oligonucleotides (23-mer, 24-mer, and 25-mer) are separated according to size. Conditions: eluent A: 0.1 M tetrabutylammonium bromide (TBuABr); eluent B: 0.1 M TBuABr, 50% acetonitrile; linear gradient of 60% to 100% B in 10 minutes, 100% B for 5 minutes, 100% to 60% B in 2 minutes; 69°C; 260 nm; 0.75 mL/minute. (Chromatogram courtesy of Transgenomic, Inc., Omaha, NE, USA.)
Under a nondenaturing temperature (50°C), the system provides gel-like separation of multiple fragments. Published reports have shown good results using this method for length polymorphism assays,10 QT-RT-PCR,11 and quality control assays of PCR products.12 The separated products can be collected and purified, making them suitable for sequencing or cloning.13
These alternative uses of the system are important to consider, because many sequencing core laboratories must face a capital expenditure to add DHPLC as a service to their facility. This expenditure could be justified if the system was used to provide other functions in addition to DHPLC analysis.
SSCP, described in Orita's watershed paper in 1989,14 has proved to be a useful tool for the detection of SNPs. It has been used successfully for detecting unknown mutations and for screening known mutations.15,16 The technique is simple and fits well in any laboratory where sequencing or genotyping is performed.
In PCR-SSCP analysis, an amplified product of a single size is denatured and prevented from reannealing to its complement by dilution, chilling, or the addition of denaturing agents such as NaOH. This forces each strand to reanneal to itself in a characteristic stable conformer. When these conformers are electrophoresed under native (nondenaturing) conditions, the mobility of each strand is affected by its conformational structure. This structure depends on the sequence of nucleotides within the strand (Fig. 6). Sequence variation within the PCR product, such as a point mutation, produces a different conformer, resulting in a change in electrophoretic mobility. By comparing the strand mobilities between samples, genomic variation can be identified.
FIGURE 6. Single-strand conformation polymorphism analysis. Each allele of a PCR amplimer forms a unique stable conformer depending on sequence. A single nucleotide change alters the folded structure of each denatured strand, with a concomitant change in electrophoretic mobility.
SSCP has traditionally been performed on standard polyacrylamide gels, using p32 (Fig. 7), silver staining, or ethidium bromide staining. An alternative is fluorescent SSCP (Fig. 8).17 Fluorescent SSCP has three major advantages over monochrome methods such as p32 or silver staining. First, PCR primers can be fluorescently labeled and the amplicons run on automated sequencers, saving time by eliminating the postrun handling of gels and eliminating the use of hazardous radiation sources and toxic silver-staining chemicals. Second, each PCR primer is labeled a different color, enabling the detection of subtle mobility shifts or shifts in which the mutant strands switch mobilities with the wild-type strands. The third major advantage is that a control can be fluorescently tagged with a third fluor and run in each lane. The control can be a commercial standard such as the GS350 size standard (PE Biosystems, Foster City, CA, USA), or for even greater sensitivity, a known wild-type sample can be PCR amplified with a third fluor. An internal control makes it possible to stagger load samples or load them in well-type combs to eliminate false readings through lane leakage. Fluorescent SSCP gels are analyzed with GeneScan (PE Biosystems) and efficiently scored with Genotyper software.
FIGURE 7. PCR-SSCP mutation detection analysis. In this 220-bp fragment, a C-->T transition in sample 7 causes a change in the conformeric structure of each strand. This change results in a dramatic mobility shift in the lower band and a subtle shift in the upper band. Gel conditions: p32; 6% acrylamide, 5% glycerol; 6 hours at 20°C. (Courtesy of M. E. Kaplan, University of Arizona, Tucson, AZ.)
FIGURE 8. Fluorescent SSCP analysis. Wild-type DNA is amplified with HEX-labeled primers (thick lines) and run in each lane as a control. Normal, heterozygous mutant and homozygous mutant samples are amplified with TET-labeled forward primer (dotted lines) and FAM-labeled reverse primer (thin lines). Under these conditions, only the forward strand was informative, as indicated by the mobility shift of the green band. Conditions: MDE 0.5X gel; run 7 hours at 10 W.
The Genomics Core Facility at the University of Utah has developed protocols for SSCP using a PE Biosystems 373 automated sequencer. These protocols are largely based on recommendations from PE Biosystems' GeneScan Reference Guide.
The PE Biosystems 310 is a single-capillary instrument that can be used successfully for SSCP.18 It has temperature control from ambient to 60°C. An autosampler loads samples, which take about 20 minutes to run. The 310 requires a new matrix for each new polymer condition. The potential throughput is 72 injections per day. For the PE Biosystems 3700, the capillary matrix is different from the 310 and does not accommodate SSCP analysis.
Although the PE Biosystems 373 does not have an active temperature controller, it can be used successfully at ambient temperature by running at low wattage. The PE Biosytems' GeneScan Reference Guide claims that fragments smaller than 300 bp can be separated in 4 hours. The Genomics Core Facility at the University of Utah has developed SSCP protocols for detecting a variety of mutations using this instrument at ambient temperature. Optimal results have been obtained by running gels for 7 hours at 10 W using Mutation Detection Enhancement (MDE) gels (FMC BioProducts, Rockland, ME, USA) and an internal control of HEX-labeled wild type. Instruments with the XL upgrade can accommodate a chiller unit, further expanding the instrument's capabilities. Running the 373 in a 4°C cold room is not recommended because condensation inside the instrument could cause considerable damage and create an electrical hazard. The potential throughput of a 36-lane instrument is 72 to 108 lanes per day.
The PE Biosystems 377 can be successfully used for SSCP at ambient temperature. However, it has the potential to run at subambient temperatures when retrofitted with a chiller device. This greatly expands the range of conditions for SSCP and allows faster runs. Any of several external cold water bath setups can be used; PE uses as an example the NESLAB Model RTE111 (NESLAB, Inc., Portsmouth, NH, USA). Minimum requirements, setup schematics, and additional information can be found on pages 7-4 to 7-9 of the GeneScan Reference Guide. The ABI Prism 377 Collection Software, version 2.1 or higher, is necessary because it has specific chiller run modules and can disengage the instrument's internal heat control system and switch to the external chiller. Running the 377 in a 4°C cold room is not recommended because condensation inside the instrument could cause considerable damage and create an electrical hazard. PE Biosytems' GeneScan Reference Guide claims that fragments smaller than 300 bp can be separated in 2 hours. The potential throughput of a 96-lane instrument is 480 lanes per day.
SSCP is most useful for amplicons 130 to 250 bp long (PE Biosystems). Sheffield et al. demonstrated that the optimal PCR fragment size for SSCP is 150 bp and that the percentage of mutations detected dropped with increasing fragment size and was compromised with very small fragments.19 Sensitivity can vary among laboratories and loci, but published reports have shown it to be between 70% and 100% for fragments in this size range.20,21
Because conformer mobility is highly sensitive to relatively minor changes in electrophoresis conditions, optimization of SSCP is necessary. The optimization for each mutation in each amplicon is empiric. The gel conditions most often manipulated are temperature of electrophoresis (from room temperature to 4°C); gel matrices (standard polyacrylamide, Long Ranger, MDE); gel concentration; and gel additives (urea, formamide, 5% to 15% glycerol). Highsmith et al. discussed various parameters affecting the sensitivity of SSCP, giving us some predictive help.22 They demonstrated that, although greater overall sensitivity can be gained by decreasing gel temperature, higher GC content correlates with greater sensitivity at room temperature.
When using SSCP to score known mutations, the optimization is often straightforward, because the goal is to find a condition that identifies that particular mutation. However, when searching for uncharacterized mutations, it is usually necessary to run several different conditions (typically two to six) to ensure identification of all sequence variations.23
After SSCP conditions have been established to detect a single known mutation in a well-characterized gene, samples can be genotyped under those conditions, and subsequent sequencing may not be necessary. However, when using SSCP for the detection of unknown mutations, a second step of sequencing is necessary to determine the nature of the variant. Because it is not possible to excise the aberrant band from the gel with automated instrumentation, the sequence must be obtained at the cost of an additional PCR reaction.
As in other types of high-throughput fragment analysis, multiplexing greatly improves cost and time efficiency. In the Genomics Core Facility at the University of Utah, the cost to PCR, run, and analyze a single gel lane with a single PCR product is $5.25, which covers reagents, labor, and instrument service charges. A multiplex of 4 amplicons reduces the per-genotype cost to less than $2.00 (Table 1). For comparison, sequencing each amplicon would cost $15.00 per amplicon (Average Lab Portrait, 1998 ABRF DNA Sequencing Research Committee Study). Multiplexing strategies include PCR coamplification,24 PCR product pooling,15 and stepped PCR primers (Fig. 9).25 The time saved through multiplexing, regardless of the strategy used, is often as important a consideration as the cost.
TABLE 1
Genomics Core Facility Cost Calculation, Showing Decreasing Cost
per Genotype With Increasing Number in the Multiplex
|
|
||||||||
| Multiplex (No. of amplicons) |
Basic Lane Cost |
Basic PCR Cost |
Total Cost per Lane |
Total Cost per Genotype |
||||
|
|
||||||||
| 1 | $4.50 | $0.75 | $5.25 | $5.25 | ||||
| 2 | $4.50 | $1.50 | $6.00 | $3.00 | ||||
| 3 | $4.50 | $2.25 | $6.75 | $2.25 | ||||
| 4 | $4.50 | $3.00 | $7.50 | $1.88 | ||||
| 5 | $4.50 | $3.75 | $8.25 | $1.65 | ||||
| 6 | $4.50 | $4.50 | $9.00 | $1.50 | ||||
|
|
||||||||
FIGURE 9. Multiplex SSCP analysis of 4 PCR amplimers. The wild-type (solid lines) control is shown, and the TET-labeled (dotted lines) forward strand is shifted to the right in mutants. Conditions: MDE 0.5X gel; run 7 hours at 10 W.
A 100-nmol synthesis of a fluorescently tagged 20-mer with standard purification costs about $80.00. Labeling both primers is suggested because a mobility shift often is seen in only one strand. Labeling both strands also allows optimization of gel conditions more rapidly, saving labor costs and instrument time. PE Biosystems recommends using FAM and TET for labeling the primers because they are of similar intensity.
To detect the smallest changes in conformer mobility, it is necessary to run internal controls on every sample. PE Biosystems' GS350 TAMRA costs about $.25 per lane. For large-scale projects, it is more cost effective to amplify a normal standard with a third fluor. Purchasing labeled primers and amplifying a half-dozen 100-µL reactions can provide enough normal standard for thousands of reactions.
SSCP can also be used to analyze samples with a mixed population of cells, such as tumor DNA, for which sequencing may prove inconclusive. In a RAS mutation, sequencing showed a suspicious background peak. SSCP provided distinct peaks that confirmed the mutation (Fig. 10), with the peak height difference showing the relative abundance of normal and mutant DNA in the sample.
FIGURE 10. Confirmation of inconclusive sequence. Wild-type DNA (thick lines) is amplified with HEX-labeled primers and run in each lane as a control. The normal sample shows one peak for the FAM-labeled strand (dotted lines) and one peak for the TET-labeled strand (thin lines). Tumor 2 shows a clear mobility shift for the TET-labeled strand. Tumor 3 was inconclusive by sequencing because of the high percentage of contaminating normal cells, but single-strand conformation polymorphism analysis reveals the presence of a mutant population. Conditions: MDE 0.53 gel; run 7 hours at 10 W.
Although sequencing remains the gold standard for mutation detection, DHPLC and SSCP should be considered when the project is large enough to offset the initial start-up costs, the project can take advantage of multiplexing, or mutations are expected to occur only rarely. These methods may be valid alternatives to sequencing when analyzing large sample sets, because the sequence of each individual within the set can be inferred if the comparative analysis offered by SSCP or DHPLC shows 100% homology between them.
Table 2 summarizes the throughput and costs associated with each method. SSCP has the advantage of being able to use equipment that is most likely already in place in the sequencing core facility. DHPLC has the advantage of having a higher throughput and a lower per-sample cost but may require an initial capital expenditure. Considerations about adding DHPLC, SSCP, or both to a sequencing core facility should involve a needs assessment of the community being served. Such an assessment can help identify which method would best serve those needs and which method would be most beneficial for the existing facility.
TABLE 2
Summary of Costs and Throughput of DHPLC and SSCP for Single Fragment
Analysis
|
|
||||||||||||
Method |
Capital Expenditurea |
Per Analysis Chargeb |
Optimal Size |
Samples per Day |
Base Pairs per Dayc |
Optimization Requirements |
||||||
|
|
||||||||||||
| DHPLC | HPLC system | $1.50 | 100-500 bp | 240 | 120,000 | Generate melting profile on | ||||||
| control sample | ||||||||||||
| SSCP | $5.25 | 130-250 bp | Run controls with 2 to 6 | |||||||||
| ABI 310 | None | 72 | 18,000 | gel conditions | ||||||||
| ABI 3732-6d | None | 108 | 27,000 | |||||||||
| ABI 377d | Chiller (optional) | 288 | 72,000 | |||||||||
|
|
||||||||||||
| DHPLC, denaturing high-performance liquid chromatography; SSCP, single-strand conformation polymorphism. |
| aAssumes automated sequencers are already in place. |
| bAnalysis charge includes reagents, labor, and instrument service contracts. |
| cBased on maximum optimal size. The number of base pairs analyzed per day would increase with multiplexing strategies. |
| dBased on three gels per day: 373 using 36-well comb and 377 using 96-well comb. |
1. Hayashi K. PCR-SSCP: a method for detection of mutations. Genet Anal Biomol E 1992;9:73-79.
2. Oefner PJ, Underhill PA. DNA mutation detection using denaturing high-performance liquid chromatography (DHPLC). In Current Protocols in Human Genetics, suppl 19. New York: Wiley & Sons, 1998:7.10.1-7.10.12.
3. Huber CG, Berti GN. Detection of partial denaturation in AT-rich DNA fragments by ion-pair reversed-phase chromatography. Anal Chem 1996;68:2959-2965.
4. Underhill PA, Jin L, Lin AA, et al. Detection of numerous Y chromosome biallelic polymorphisms by denaturing high-performance liquid chromatography. Genome Res 1997;7:996-1005.
5. Huber CG. Micropellicular stationary phases for high-performance liquid chromatography of double-stranded DNA. J Chromatogr A 1998;806:1-28.
6. Liu W, Smith DI, Rechtzigel KJ, Thibodeau SN, James CD. Denaturing high performance liquid chromatography (DHPLC) used in the detection of germline and somatic mutations. Nucleic Acids Res 1998;26:1396-1400.
7. O'Donovan MC, Oefner PJ, Roberts SC, et al. Blind analysis of denaturing high-performance liquid chromatography as a tool for mutation detection. Genomics 1998;52:44-49.
8. Bonner MR, Stone AC, Taylor P, Hammer MF. Increasing the throughput of DHPLC for detecting mutations in DNA. Am J Hum Genet Suppl 1998;63:1300.
9. Huber CG, Oefner PJ, Bonn GK. High-resolution liquid chromatography of oligonucleotides on nonporous alkylated styrene-divinylbenzene copolymers. Anal Biochem 1993;212:351-358.
10. Marino MA, Devaney JM, Smith JK, Girard JE. Sequencing using capillary electrophoresis of short tandem repeat alleles separated and purified by high performance liquid chromatography. Electrophoresis 1998;19:108-118.
11. Hayward-Lester A, Oefner PJ, Doris PA. Rapid quantification of gene expression by competitive RT-PCR and ion-pair reversed-phase HPLC. Biotechniques 1996;20:250-257.
12. Huber CG, Oefner PJ, Preuss E, Bonn GK. High-resolution liquid chromatography of DNA fragments on non-porous poly(styrene-divinylbenzene) particles. Nucleic Acids Res 1993;21:1061-1066.
13. Hecker KH, Turpie B, Kuklin A. Optimization of cloning efficacy by pre-cloning DNA fragment analysis. Biotechniques 1999;26:216-222.
14. Orita M, Suzuki Y, Sekiya T, Hayashi K. Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 1989;5:874-879.
15. Winterpacht A, Hilbert K, Schwarze U, Zabel B. Non-radioactive multiplex-SSCP analysis: detection of a new type II procollagen gene (COL2A1) mutation. Hum Genet 1995;95:437-439.
16. Wenz HM, Baumheuter S, Ramachandra S, Worwood M. A rapid automated SSCP multiplex capillary electrophoresis protocol that detects the two common mutations implicated in hereditary hemochromatosis (HH). Hum Genet 1999;104:29-35.
17. Makino R, Yazyu H, Kishimoto Y, Sekiya T, Hayashi K. F-SSCP: fluorescence-based polymerase chain reaction-single-strand conformation polymorphism (PCR-SSCP) analysis. PCR Methods Appl 1992;2:10-13.
18. Inazuka M, Wenz HM, Sakabe M, Tahira T, Hayashi K. A streamlined mutation detection system; multicolor post-PCR fluorescence labeling and single-strand conformational polymorphism analysis by capillary electrophoresis. Genome Res 1997;11:1094-103.
19. Sheffield VC, Beck JS, Kwitek AE, Sandstrom DW, Stone EM. The sensitivity of single-strand conformation polymorphism analysis for the detection of single base substitutions. Genomics 1993;16:325-332.
20. Hayashi K, Yandell DW. How sensitive is PCR-SSCP? Hum Mutat 1993;2:404-414.
21. Ellison J, Dean M, Goldman D. Efficacy of fluorescence-based PCR-SSCP for detection of point mutations. Biotechniques 1993;15:684-691.
22. Highsmith WE Jr, Nataraj AJ, Jin Q, O'Connor JM, El-Naby SH, Kusukawa N, Garner MM. Use of DNA toolbox for the characterization of mutation scanning methods. II: Evaluation of single-strand conformation polymorphism analysis. Electrophoresis 1999;20:1195-1203.
23. Glavac D, Dean M. Optimization of the single-strand conformation polymorphism (SSCP) technique for detection of point mutations. Hum Mutat 1993;2:404-414.
24. Ozcelik H, Andrulis I. Multiplex PCR-SSCP for simultaneous screening for mutations in several exons of p53. Biotechniques 1995;18:742-744.
25. Ballard L, Walker A, Scholl R. Multiplex fluorescent SSCP using stepped PCR primers. Biotechniques (submitted).