| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Life Sciences Research, LGC Ltd., Queens Road, Teddington, Middlesex, TW11 0LY, United Kingdom.
a Author for correspondence. Fax 44-020-89432767;
hcp{at}lgc.co.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Two standardized assays designed to minimize the introduction of non-thermal-cycler-dependent variations were evaluated by 18 laboratories in the United Kingdom, using 33 thermal cyclers of various makes and models. We used a single-product (590 bp) PCR, established in our laboratory as a robust and specific reaction. The second reaction, a multiproduct random amplified polymorphic DNA (RAPD) PCR, was known to be more susceptible to small changes in block temperature and was therefore considered a way of assessing block uniformity with respect to temperature. Assay repeatability data were analyzed with respect to temperature calibration status, the type of temperature control mechanism, thermal cycler age, and the presence of oil overlay or heated lid systems.
Results: All (100%) of the laboratories produced the correct target for the single-product PCR assay, although substantial variation in yield in replicate reactions was observed in 9.4% of these. The RAPD reaction generated results that varied extensively both within the same block and between different thermal cyclers. For eight replicates of a positive sample, 88% intrablock repeatability was demonstrated in calibrated thermal cyclers, which decreased to 63% in noncalibrated instruments.
Conclusions: Irrespective of the make and model of thermal cycler, temperature-calibrated instruments consistently generated more repeatable RAPD data than noncalibrated instruments. Guidelines are offered on optimizing and monitoring thermal cycler performance.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thermal cyclers are the programmable heating blocks that control and
maintain the temperature of the sample through the three
temperature-dependent stages that constitute a single cycle of PCR:
template denaturation (
95 °C); primer annealing
(35–65 °C); and primer extension (72 °C). These temperatures
are cycled up to 40 times to obtain amplification of the DNA target. If
the optimum programmed temperatures are not met because the block
overshoots or undershoots them or because they are not consistent
across the block, reaction specificity and sensitivity can be
compromised, leading to the consequences listed above and ultimately to
the false interpretation and reporting of data.
The aim of this study, therefore, was to investigate the effects of thermal cycler performance on PCR reproducibility and repeatability, using two standardized assays designed to minimize the introduction of non-thermal-cycler-dependent variations. The first was a single-product (590 bp) PCR, using primers designed against multicopy ribosomal genes (2). This was established in our laboratory as a robust and specific reaction that consistently produced the correct product when performed under a broad range of conditions. The second test assay was a multiproduct random amplified polymorphic DNA (RAPD)1 reaction. This was relatively nonspecific because of the short 10-base primer and low annealing temperature used, and was shown, in our hands, to be sensitive to slight variations in both annealing temperature and temperature ramp rate (specifically the temperature transition rate from annealing to extension). These characteristics of the RAPD reaction rendered it particularly suitable to assess well-to-well uniformity across the block of a thermal cycler (intrablock repeatability) and to highlight block-to-block variations between different thermal cyclers (interblock reproducibility). The temperature calibration status of the thermal cyclers was also established, and its significance was investigated. For the purposes of this study, thermal cycler temperature calibration was defined as the uniformity of the block temperature with an accuracy ± 1 °C of the programmed temperature across the entire block, as assessed by a recognized calibration service in the year before this study.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
single-product pcr
The single-product PCR reaction mixtures consisted of 1x PCR
buffer (50 mM KCl, 10 mM Tris-HCl, pH 8.3, and 0.01 g/L
gelatin), 2 mM MgCl2 (both from PE Applied
Biosystems), 200 µM dNTPs (Pharmacia Biotech), 2 µM each
primer (U1, 5'-CAGCCGCCGCGGTAATTC-3'; U2, 5'-CCGTCAATTCCTTTGAGTTT-3';
synthesized and HPLC purified by Genosys Biotechnologies), 0.125 ng of
DNA (Saccharomyces cerevisiae; Sigma), 0.02 U/µL
Taq DNA polymerase (Pharmacia Biotech), and doubly distilled
H2O. Samples were amplified as follows: 5
min at 94 °C (initial denaturation); 30 cycles of 90 s at
94 °C, 15 s at 50 °C, 120 s at 72 °C; and 7 min at
72 °C (final extension), followed by a final hold at 4 °C.
multiproduct rapd
The RAPD reactions consisted of 1x PCR buffer (50 mM KCl, 10 mM
Tris-HCl, pH 8.3, and 0.01 g/L gelatin), 1.5 mM
MgCl2, 150 µM dNTPs, 3 µM primer (OPA-12,
5'-TCGGCGATAG-3'; Operon Technologies), 0.125 ng of DNA, 0.03 U/µL
Taq DNA polymerase, and doubly distilled
H2O. Samples were amplified as follows: 7 min at
94 °C (initial denaturation); 30 cycles of 60 s at 94 °C,
90 s at 40 °C, 120 s at 72 °C; and 10 min at 72 °C
(final extension), followed by a final hold at 4 °C.
standardization of test assays
Every effort was made to minimize the introduction of variation
not caused by thermal cycler performance. This included providing
standardized sample DNA, reagents and reaction tubes; minimizing the
pipetting undertaken by participants; supplying a detailed protocol;
and requesting that only experienced PCR users should take part in the
study. The standardized sample DNA and reagents were prepared at LGC
(Teddington) Ltd. in two master mixtures per assay, before excess
volumes were aliquoted out and dispatched to participants. One master
mixture contained 1x PCR buffer, MgCl2, dNTPs,
primers, DNA, and doubly distilled H2O. The other
contained Taq DNA polymerase only. The isolation of
Taq DNA polymerase from the DNA template prevented the
occurrence of any nonspecific reactions, whereas the simple addition of
one reagent mixture to another by the participant would minimize the
introduction of pipetting errors. Because differences in the thickness
of tubes from different suppliers may affect the temperature that the
sample reaches, standardized thin-walled microtubes (Alpha Lab
Supplies) of appropriate sizes were also supplied. In addition to the
usual precautions of performing PCR tests in a clean environment, it
was also requested that filtered tips or positive-displacement pipettes
be used for each liquid transfer to avoid contaminating the test assays
with foreign DNA.
The final detailed protocol was produced as a result of comments by participants on a draft copy, circulated for discussion before the study. The protocol stated that all liquid transfers were to be carried out on ice. To ensure complete homogenization of solutions before amplification, prepared reactions were briefly vortex-mixed and the contents centrifuged before being placed in the thermal cycler. To make direct comparisons between thermal cyclers, PCR reactions were positioned in designated wells within the thermal cycler block. Variation attributable to the different block layouts exhibited by different makes and models of thermal cycler was minimized by specifically designating positions to maintain a similar number of corner (n = 4), edge (n = 4; 2 negative and 2 positive reactions) and internal (n = 2) well positions. All assay components were dispatched to participating laboratories by next-day courier under dry ice, with instructions for reagents to be stored at -20 °C upon arrival. A 2-week turnaround time was given for completion of the test assays.
participants of the interlaboratory study
Any laboratory in the United Kingdom that was an established and
regular PCR user was invited to take part in this study, either by
personal invitation or through a notice placed in the Diagnostics Club
newsletter Exchange. A total of 18 laboratories,
representing many different sectors of DNA analysis, participated in
the study. The test assays were carried out on 33 thermal cyclers of
differing model and age and from various manufacturers.
pcr product analysis and data interpretation
Aliquots (10 µL) of the products from both assays were analyzed
by electrophoresis on 2% agarose in 1x Tris-borate-EDTA buffer (10x
stock: 8.9 mol/L Tris-HCl, pH 8.0, 8.9 mol/L boric acid, 20 mmol/L
EDTA) containing 0.5 µg/mL ethidium bromide, and visualized under
ultraviolet illumination. The remaining products and a gel
photograph of the results were returned to LGC. After arrival at LGC,
the products from both assays were reanalyzed by electrophoresis as
described above. On this occasion, to facilitate direct block-to-block
comparisons, all of the reactions amplified in specific block positions
from all thermal cyclers were analyzed on the same gel.
In one case, the operator knew that a particular thermal cycler consistently failed to reach the programmed temperatures by a defined amount, and the programmed temperatures were therefore routinely adjusted to correct for this error. Data were generated using this piece of equipment both at the adjusted and stated temperatures.
All of the results were analyzed and interpreted independently by two individuals. No image analysis equipment was used. When the results were analyzed, two sets of RAPD results were not included: those generated by the nonadjusted thermal cycler mentioned above and one set of results that was not returned by the participating laboratory (a total of 31 instruments analyzed). Analysis of the single-product PCR also excluded data from the nonadjusted thermal cycler (a total of 32 instruments analyzed).
pcr block well performance: statistical treatment and
interpretation of "nonrepeatable" counts
The counting scheme used had some properties that may lead to
biased indications of consistency for particular positions: There were
four corner positions, two edge positions, and two interior positions,
potentially biasing counts toward "corner" positions. However,
calculation on the basis of binomial probabilities showed that mean
counts were unbiased where a single profile predominated. In addition,
no total count could be equal to 1; the possible values were 0 and
2–8. The distribution was likely to be, at best, a censored binomial
distribution, making formal confidence intervals hard to estimate
accurately. Monte Carlo simulations using the observed probabilities
showed that for match counts >50% of the possible number of profiles,
the distribution followed binomial expectations closely. Confidence
intervals, where shown, were accordingly calculated at the 95% level,
using the methods described by Howarth (3) and assuming
independent profiles within each group.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Results generated at comparable well positions on different thermal
cyclers were also compared. For example, the results obtained from well
position 6 are illustrated in Fig. 1
. As expected, the single-product PCR was also highly
reproducible between different thermal cyclers, with all instruments
investigated (except instrument 19) generating the correctly sized
product. The results generated from the unadjusted and adjusted
amplifications (instruments 19 and 18, respectively) were of particular
interest because they clearly illustrated the importance of temperature
calibration to generate positive results. Six amplification failures
were recorded when reactions were performed using the unadjusted
instrument. However, when that instrument was adjusted to compensate
for the known temperature bias, all replicate PCR reactions generated
positive results. We do not, however, recommend operator-mediated
thermal cycler temperature adjustments as a suitable course of action
to alleviate temperature bias.
|
multiproduct rapd assay
Interpretation of the RAPD results was based on the presence or
absence of specific bands in each profile. A diagram of the maximum
number and positions of the amplified RAPD products is illustrated
schematically in Fig. 2
.
|
block-to-block variation
Block-to-block variation was assessed by dividing the profile into
four separate categories (shown in Fig. 2
): a "core" product (940
bp); six high-molecular weight (HMW) products; five mid-molecular
weight products; and a single low-molecular weight (LMW)
product. Generation of the remaining amplification products was less
repeatable. These, indicated in Fig. 2
by dashed lines, were considered
to be on the borderline of detection and, as a consequence, were not
used to generate any data for profile analysis.
In general, RAPD reproducibility was found to vary markedly between
different thermal cyclers (typical examples are given in Fig. 3
), highlighting the variation in results that can occur when set
thermal profiles are transferred between different thermal cycler
instruments. It should be noted, however, that in contrast to the
single-product PCR results, the profiles generated using the
nonadjusted and temperature-adjusted thermal cycler (Fig. 2
, lanes 19
and 18, respectively) showed apparent similarities. This observation is
presumably attributable to the less stringent nature of the RAPD
analyses. An analysis of the average number of amplification products
generated within each of the above categories is illustrated in Table 1
. One fragment, the core product, appeared to be more
reproducible than the others. This was present in all RAPD profiles
analyzed, with the exception of instruments 1 (total amplification
failure at position 1), 15 (failure at positions 3 and 6; very faint
product for remaining positions) and 27 (total amplification failure at
position 8). Generation of the remaining fragments selected for
analysis was, on average, less reproducible both within and between
thermal cyclers. A direct comparison between temperature-calibrated and
noncalibrated instruments demonstrated no significant difference in the
average number of bands generated for both categories of instrument.
There was, however, a clear difference in the average number of HMW and
LMW bands produced in thermal cyclers using different temperature
control mechanisms. The external simulated tube control (ESTC) thermal
cyclers produced, on average, fewer HMW bands and more frequently
produced the LMW band. In contrast, the internal computer-simulated
control (ICSC) method typically produced a higher number of HMW bands
and produced the LMW band less frequently.
|
|
well-to-well variation
The RAPD profiles were further analyzed by comparing the total
number of repeatable profiles obtained from an individual block. For
this purpose, a repeatable profile was defined as the most frequently
occurring profile out of the eight produced by each individual thermal
cycler. For example, if a single thermal cycler generated two similar
profiles and six dissimilar profiles, then that block was scored as
producing two repeatable profiles or six nonrepeatable profiles.
In general, RAPD repeatability was found to vary markedly between
different tube positions within the same thermal cycler block; typical
examples are illustrated in Fig. 4
. The effects of calibration status, thermal cycler age,
temperature control mechanism, and sample evaporation control mechanism
on the average number of repeatable profiles generated per thermal
cycler was further investigated, and the results obtained are
summarized in Table 2
. On average, calibrated instruments were found to generate a
greater number of repeatable RAPD profiles compared with noncalibrated
instruments. For calibrated instruments, out of the total number of
profiles generated per thermal cycler, approximately seven of eight
were repeatable. In contrast, for the noncalibrated equipment, this
value was reduced to five of eight. The older thermal cyclers that
participated in this study (those >3 years of age), appeared, on
average, to produce less repeatable profiles than their younger
counterparts, but the significance of the age/repeatability
relationship was marginal (P = 0.08 for linear
regression of age against number of repeatable profiles). An
investigation of the relationship between the thermal cycler
temperature control mechanism and data repeatability indicated that
ICSC instruments were superior, generating a median of 8 and mean of 7
(of 8) repeatable profiles, compared with means of 5.3 and 4.5 for
block control (BC) and ESTC instruments, respectively. Furthermore,
heated-lid thermal cyclers were found to generate more repeatable data,
displaying a median of 8 and mean of 6.6 repeatable profiles compared
with the median of 5 and mean of 4.8 generated by instruments that
required an oil overlay.
|
|
Instrument temperature calibration is an independent variable, allowing valuable observations to be derived with respect to well position and profile repeatability. However, because of the size of the study, the effects on repeatability of the other variables analyzed were of less statistical significance as these were inherently interconnected. For example, in the comparison between different temperature control mechanisms, five of the eight ICSC instruments (55.5% of the total number of calibrated instruments in the study) were calibrated. Furthermore, all eight ICSC instruments possessed heated lids. However, the ICSC instruments were of a variety of ages: three were older than 4 years, two older than 2 years, two older than 1 year, and 1 of unknown age. Temperature-calibrated instruments were split equally between the heated-lid and oil overlay categories. The heated-lid thermal cyclers consisted of all eight ICSC instruments, four BC, and one ESTC monitoring system. Once again, a range of ages of thermal cyclers was represented in the heated-lid category: four were older than 4 years, one older than 3 years, four older than 2 years, three older than 1 year, and two of unknown age. Moreover, older thermal cyclers tended to have oil overlays and BC as their mechanism of temperature control.
The areas of the thermal cycler blocks that most commonly failed to
achieve a uniform temperature were identified by analysis of the
proportion of nonrepeatable profiles generated at those positions
[(number of repeats observed)/(number of wells)]. Results obtained
demonstrated that the internal sites, on average, produced the highest
number of nonrepeatable results: 12.5 profiles out of 31, compared with
9.3 and 8.5 for the corner and edge positions, respectively (Fig. 5
).
|
The number of non-repeats produced per profile generated was also
examined with respect to the temperature calibration status of the
thermal cyclers (Fig. 6
). In this study, calibrated thermal cyclers appeared to
generate significantly fewer nonrepeatable RAPD profiles than
noncalibrated instruments. Interestingly, the number of non-repeats at
corner wells was the same irrespective of calibration status, whereas
at the edge and internal positions, the number of nonrepeatable
positions was reduced from 0.35 to 0.1 and from 0.5 to 0.1 non-repeats
per profile when a comparison was made between noncalibrated and
calibrated instruments, respectively.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
thermal cyclers
Inherent differences in the mechanisms of thermal cycler
performance could account for some of the variation observed between
different makes and models of thermal cycler. For this study, the
thermal profile specified in the protocol stipulated that the ramp
rates should be set at maximum. The technical specifications of the
instruments used in this study, however, showed that the maximum ramp
rates achievable varied from 1 to 3 °C/s, depending on the thermal
cycler. Furthermore, the RAPD assay is known to be sensitive to small
fluctuations in temperature. Consequently, it was important to consider
the influence of the various temperature control mechanisms used in the
design of the thermal cyclers on the actual temperatures reached by
individual samples and their impact on the resulting profiles generated
(1). The different mechanisms used by the thermal cyclers in
this study to monitor and control block and/or sample temperature were
highlighted and are described below.
For these reasons, it is understood that the requirement that all instruments use the same thermal profile introduced a certain bias into the temperatures obtained by the samples themselves. It is therefore essential to stress that, for the purposes of this study, there was no "correct" profile for the RAPD reaction and that analysis of RAPD reproducibility would primarily highlight the implications of transferring a set thermal protocol from one thermal cycler to the next. Other differences in thermal cyclers include (a) the use of a heated lid or mineral oil overlay to minimize evaporation of the sample; (b) the thermal cycler ramp rate; (c) the reaction tube size; and (d) the presence or absence of a cooling facility, i.e., the capacity of the thermal cycler to attain subambient temperatures. All of these variables emphasize the potential risks with regard to test reproducibility that may be associated with transferring a set thermal profile from one type of thermal cycler to another.
calibration
Calibration is the process of establishing the response of an
instrument to a set of conditions. In the case of thermal cycler
calibration, a programmed temperature is allowed to equilibrate before
the actual temperature is measured with a temperature probe placed in a
well. Although initial responsibility for calibration of the thermal
cycler lies with the manufacturer or vendor, sale or installment of the
equipment transfers the responsibility of calibration to the testing
laboratory; however, it is perfectly in order and perhaps prudent for
the manufacturer to offer calibration recommendations. There probably
is not an analytical or research situation where calibration is not
strongly recommended and in the majority of cases should be enforced.
If results need to be generated using a different thermal cycler
instrument, the validity of the data could be open to question if the
instrument responds in a different manner to a set of transferred
conditions. Information regarding calibration services (definition of
calibration with respect to accuracy and uniformity, the number of well
positions tested, the range of temperatures tested, and the nature of
postcalibration instrument adjustments) can be obtained from thermal
cycler manufacturers.
interlaboratory trials
One of the first reported multicenter blinded PCR proficiency
trials appeared in 1991 (4). The authors’ final
recommendations included standardization of reagents and procedures and
ongoing participation in blinded proficiency testing (PT) schemes of
laboratories offering PCR tests. Since then, many other trials have
been published. Although these have been predominately in the clinical
field, there recently have been a growing number of reports in
nonclinical sectors (5). A review of the literature revealed
that an interlaboratory study on the performance of thermal cyclers had
not been undertaken previously. It is clear, however, that there
currently is some concern over the quality of the PCR-based data being
produced and that interlaboratory studies are seen by some as an
efficient way of assessing variability of results. Therefore, we
present a short summary of some interlaboratory studies undertaken to
study DNA technologies and some comments highlighting the possible
effects of thermal cycler performance.
Neumaier et al. (6) offered eminently sensible guidelines on the application of PCR in diagnostics and emphasized the need for standardization and definition of laboratory organization. However, the authors were concerned that because the methodology was in constant change, no general standards could be defined. They identified uniform temperature transition as an important aspect of a successful amplification and specifically noted that the heat conduction in the heating block must be controlled regularly to avoid false-negative results. They made no mention of advising on thermal cycler calibration. A European interlaboratory study investigating the reproducibility of RAPD, amplified fragment length polymorphism, and simple sequence repeat markers found that RAPDs were difficult to reproduce even when standardized reaction mixtures were used (5). The standard thermal profile was obtained by participants copying a thermal profile printed from the PCR instrument of the distributor (it is unclear whether this thermal profile was verified or actually achieved by the participants’ thermal cyclers). In spite of this, significant variation in the RAPD profiles obtained was reported (5). In another interlaboratory study, participants used their own PCR technique to detect Toxoplasma gondii (7). Detection limits were found to vary, and false-positive and false-negative results were reported. The authors concluded that until an appropriate quality assurance scheme was established, they would recommend that PCR be used in association with other laboratory methods. They appreciated the fact that accuracy of the thermal cycler instruments may have contributed to the observed discrepancies, and they also emphasized the urgent need for the standardization of PCR protocols and the possible accreditation of diagnostic laboratories. The authors further stressed that the differences in PCR design and potential discrepancies in performance of PCR highlighted in the study underlined the caution required in applying findings (particularly PCR sensitivity and specificity) from other published clinical studies using different PCR methodologies. Weber et al. (8) conducted a study for the detection of the human polyomavirus JC (JCV), using PCR. Five of the six participants reported a detection limit of 1–10 JCV copies per 10 µL, and the remaining participant reported a detection limit of 100 000 JCV copies per 10 µL. This marked difference in sensitivity was ascribed to nonspecified "thermocycler problems".
enhancing thermal cycler performance and confidence in data
This interlaboratory study has increased awareness among the
participants of the importance of validation issues in PCR-based
analyses, emphasized the importance of instrument calibration, and
raised interest in the possible use of regular PT schemes in DNA
analysis. Such schemes could offer several benefits, including external
recognition, increased competitiveness, and improved quality of data.
They also give an independent assessment of performance and allow a
comparison of results between laboratories. Furthermore, tight internal
quality control can lead to bias, and an external check may be
desirable to compare the precision of one laboratory with another. In
other words, although a laboratory may be producing consistent and
repeatable results, there is no guarantee that these results are
actually correct. This can be verified only after comparison with other
laboratories. Of all of the participants in this interlaboratory study,
only four had previously participated in some form of PT scheme. The
majority of participants agreed that thermal cycler calibration and
temperature monitoring are important issues for the generation of
reliable and repeatable data.
There are numerous steps that can be taken to maintain or improve thermal cycler performance. Yearly temperature calibration by a reputable thermal cycler manufacturer is considered a minimum requirement. Performance can be further assured throughout the year by in-house calibration using a sensitive test, such as a RAPD assay. Monthly runs of such a biological assay should highlight any variation in performance. One advantage of a temperature-sensitive biological control resides in the fact that it monitors dynamic temperature uniformity, whereas a temperature probe-type calibration usually monitors only static temperature uniformity. In addition, the number of wells monitored limits temperature probe calibration. A biological control could be analyzed periodically in all block positions. It must, however, be stressed that use of a robust assay that is not temperature sensitive will lead the laboratory into a false sense of security and will be of little value. Simple housekeeping duties such as closing the lid of a thermal cycler when not in use and keeping the wells clean by the use of cotton balls and alcohol will ensure that there is a good fit between the block and the tubes, thereby maximizing heat transfer. The use of internal PCR controls is now becoming increasingly popular. These have the benefit of controlling for false-negative reactions at each well position. Such controls can be incorporated into assays in the form of competitive reactions, where the sample and control targets share common primer pairs, or in a noncompetitive reaction, where both targets are amplified with unique primer pairs. Finally, additional confidence in the data generated can be obtained by participating in interlaboratory studies or PT trials. Such an undertaking will allow independent assessment or benchmarking of performance. These schemes have become much more common in recent years. Where comparability of data is required, such multicenter collaborative trials can offer a way of improving the reproducibility of results by offering continual monitoring, standardization, and troubleshooting opportunities.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. M. Collery, D. S. Smyth, J. M. Twohig, A. C. Shore, D. C. Coleman, and C. J. Smyth Molecular typing of nasal carriage isolates of Staphylococcus aureus from an Irish university student population based on toxin gene PCR, agr locus types and multiple locus, variable number tandem repeat analysis J. Med. Microbiol., March 1, 2008; 57(3): 348 - 358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Orlando, P. Verderio, R. Maatman, J. Danneberg, S. Ramsden, M. Neumaier, D. Taruscio, V. Falbo, R. Jansen, C. Casini-Raggi, et al. EQUAL-qual: A European Program for External Quality Assessment of Genomic DNA Extraction and PCR Amplification Clin. Chem., July 1, 2007; 53(7): 1349 - 1357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Schoder, A. Schmalwieser, G. Schauberger, J. Hoorfar, M. Kuhn, and M. Wagner Novel Approach for Assessing Performance of PCR Cyclers Used for Diagnostic Testing J. Clin. Microbiol., June 1, 2005; 43(6): 2724 - 2728. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Birch, C. A. English, M. Burns, and J. T. Keer Generic Scheme for Independent Performance Assessment in the Molecular Biology Laboratory Clin. Chem., September 1, 2004; 50(9): 1553 - 1559. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Schoder, A. Schmalwieser, G. Schauberger, M. Kuhn, J. Hoorfar, and M. Wagner Physical Characteristics of Six New Thermocyclers Clin. Chem., June 1, 2003; 49(6): 960 - 963. [Full Text] [PDF]
|
|
|
|
|
|
C. C. Raggi, P. Pinzani, A. Paradiso, M. Pazzagli, and C. Orlando External Quality Assurance Program for PCR Amplification of Genomic DNA: An Italian Experience Clin. Chem., May 1, 2003; 49(5): 782 - 791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Rolland, L. Mereghetti, S. Watt, S. Chatellier, and R. Quentin tRNA gene clusters at the 3' end of rRNA operons are specific to virulent subgroups of Streptococcus agalactiae strains, as demonstrated by molecular differential analysis at the population level Microbiology, May 1, 2002; 148(5): 1493 - 1499. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
