Real-time Gene Expression Analysis as Monitoring Tool for Production of Recombinant Proteins

Sabine Knappe1, Christiana Cordes2
1 TU Dresden, Germany; 2 Hochschule Anhalt (FH)

Abstract

In recent years numerous pharmaceuticals have been developed from biotechnologically produced proteins. In the presented study we use Escherichia coli DH5alpha with the GFP mutant gene for T-Sapphire as a model. This work describes a method to quantify gene expression of the product with real-time PCR and its potential use as a monitoring tool. Efficient total RNA purification and cDNA synthesis are one of the essential steps in real-time PCR quantification. Therefore we tested various RNA purification and reverse transcription systems for their influences on the efficiency and reproducibilityof real-time PCR. To have the possibility of comparing different experiments, normalization with housekeeping genes (HKGs) is the method of choice (relative quantification). Therefore we tested the expression of ten HKGs from various pathways for normalization. Three candidates were determined by the procedure of Vandesompele et al. 2002 (gene stability measure M, normalization factor NF). With the developed methods fermentation (batch and feed-batch) under different conditions were monitored. This work was part of the project „Monitoring and Control of Bioprocesses for heterologous protein production in Escherichia coli“ and supported by BMBF (FKZ 1710C04) running from November 2004 until January 2008.


Back to Data Analysis: qPCR BioStatistics & BioInformatics

Quantitative real time-PCR-assay for the analysis of gene-specific human influenza A virus transcription & replication dynamics

Antje Lagoda1, Diana Hoffmann2, Diana Vester1, Claudius Seitz2, Yvonne Genzel2, Udo Reichl1,2
1 Chair of Bioprocess Engineering: Otto-von-Guericke University, Magdeburg, Germany; 2 Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany

Abstract

For design and optimization of influenza virus production in mammalian cells details on human influenza A virus transcription and replication dynamics are essential. In addition to the characterization of cell growth and product formation in bioreactors, the use of mathematical models describing relevant aspects of virus replication is advantageous. In a previously developed, structured mathematical model for influenza A virus replication in Madin-Darby canine kidney (MDCK) cells [1] there was a considerable lack of quantitative experimental data. In particular, information regarding dynamics and control of the synthesis of each class of intracellular viral RNA, namely vRNA(-), cRNA(+) and vmRNA(+) was missing. Therefore, polarity specific primers were used in reverse transcription to quantify corresponding RNA species by real time-PCR (RRT-PCR). Here, we present results for establishment and validation of a RRT-PCR-assay for different segments of human influenza A/PR/8/34 (HA, NA, M, NS). As part of the validation protocol repeatability, reproducibility, specificity of the polarity specific primers and sensitivity of this assay was checked using dilution series of RNA reference standards. This validated assay was then used for monitoring vmRNA(+), cRNA(+) and vRNA(-) during infection of adherent MDCK cells. Determination of the intracellular RNA concentrations showed significant differences in viral transcription and replication dynamics. Based on the high quality of quantitative data obtained, validation of mathematical models describing virus-host cell interactions is facilitated significantly. [1] Sidorenko, Y. and Reichl, U. (2004): Structured model of influenza virus replication in MDCK cells, Biotechnology and Bioengineering, 88(1), 1-14


Back to Data Analysis: qPCR BioStatistics & BioInformatics

qPCR: Application for real-time PCR data management and analysis

Stephan Pabinger1, Rene Snajder1, Robert Rader1, Heiko Eichhorn2, Zlatko Trajanoski1, Gerhard G. Thallinger1
1 Institute for Genomics and Bioinformatics, Graz University of Technology, Austria; 2 Development Anti-Infectives Microbiology, Sandoz GmbH, Kundl, Austria

Abstract

We have developed QPCR, a web application for the analysis of data from qPCR experiments that integrates the complete analysis workflow including Cq and efficiency calculations, normalization, statistical analysis, and visualization. The application comprises parsers to automatically import data created by qPCR devices. Supported are files generated by the Applied Biosystems and Roche Lightcycler instruments software as well as a generic CSV file format. QPCR provides a graphical representation of the plate layout displaying sample, detector, and status information of each well on a grid. By selecting wells in the grid, charts of amplification and dissociation curves are displayed. Furthermore every property of a stored plate is fully editable allowing adjustment of data for later analysis. QPCR currently includes four algorithms for the calculation of the Cq value and six methods for the estimation of amplification efficiency values as described by Guescini et al. (2008), Ostermeier et al. (2003), Ramakers et al.(2003), Rutledge et al. (2004), Rutledge et al. (2008), Wilhelm et al. (2003), and Zhao et al. (2005). All algorithms and data file parsers used by the application are integrated through a plug-in mechanism, which allows simple extensions to support additional qPCR data formats and analysis methods. Further analysis steps include technical and biological replicate handling, incorporation of sample or target specific efficiency, normalization using single or multiple reference genes, inter-run calibration, and fold change calculation (see Hellemans et al. (2007)). In addition to the calculation of well-based efficiencies, efficiency can be determined based on serial dilution series using a graphical interface for the parameterization. QPCR includes standard and permutation based statistical tests to evaluate differences between samples groups and supports proper error propagation throughout all analysis steps. It provides a quality control method to evaluate applicability of the reference genes used for normalization (gene stability values and coefficients of variation). Results of all analysis steps are graphically displayed and can be exported as a data file to be used in external programs. The generated charts are highly customizable and are designed to be usable in publications without further manipulation. The web application includes an authentication and authorization system ensuring confidentiality and data security. Since 2007 the application has been deployed at several institutes and has been extensively tested and adapted to user requirements.

Demo version


Back to Data Analysis: qPCR BioStatistics & BioInformatics

Quality Control in qPCR

Mario Cunha, Luis Martins, Carmo Ornelas
Clinical Pathology, Lab. Virology, Portuguese Institute of Oncology, Lisbon, Portugal

Abstract

With the introduction of Real Time PCR (qPCR) in most clinical laboratories, we have been able to combine the kinetics of PCR with fluorescent probes, in order to monitor the PCR product in Real Time. Besides the excellent sensibility and specificity, qPCR is easy to execute and has a low risk of contamination. With all these capabilities, it has become an excellent alternative to traditional methods like cell culture and immunoassays. When a new assay is implemented in a clinical laboratory, precise and exact results are an absolute requirement, two of the main characteristics of Quality Control. Since qPCR has been introduced very recently in the clinical laboratory, we must normalize the steps we have to take regarding its validation and implementation in routine. The Accreditation procedures (ISO 15189) demand validation of every assay implemented. Despite the emission, by Portuguese Institute of Accreditation (IPAC), of some guidelines, these lack some of the important steps to validation. However, there are scientific articles and international guides that describe the necessary steps to take, in order to implement qPCR assays in the clinical laboratory. Once validated and implemented, the qPCR assays must be monitored routinely in order to check for shifts/deviations from the parameters set initially. The current methodology is the Internal Quality Control (IQA) and External Quality Control (EQA). In the IQA procedure, we must introduce a Non Template Control (NTC) in every run, with or without sample replicas, using one or several Positive Controls, expressed in Cq, Copies/ml or IU/ml. The values can be monitored in Quality Control Charts (MultiQC). We can also monitor the regression parameters, establishing limits for the Slope, Y Intercept, Detection/Quantification Limit and Efficiency. Concerning EQA, the aim of this methodology is to evaluate the Laboratory’s capacity to produce good quality results. For every qPCR assay that its implemented, we must participate in an EQA program. In Virology there are several programs available, namely QCMD, NEQAS, and Instand. Both results, from IQA and EQA, are used to measure the performance of the laboratory through the determination of Total Error and Sigma. For the Total Error, we apply the following formula: TE=|Bias|+ZxCV% The Bias can be obtained when we compare our EQA results with those obtained by other laboratories; CV% can be measured by our Positive Controls, in Quality Control Charts; and Z is constant: 1,65 for CI of 95%. If we have established an Allowed Total Error (TEa), we can determine the Sigma of our qPCR assay: Sigma = (TEa – |Bias|)/CV%. The Sigma has a scale of evaluation: 0 is bad, 6 or above is perfect.


Back to Data Analysis: qPCR BioStatistics & BioInformatics

DNA Melting and Mathematics

Bob Palais1, Carl Wittwer2
1 Mathematics Department, University of Utah; 2 Pathology Departments, University of Utah

Abstract

High resolution DNA melting is a simple closed-tube method of genotyping and variant scanning. However, advanced analysis methods are crucial for accurate design and interpretation. Mathematical modeling allows melting prediction by nearest neighbor thermodynamics. Analysis needs including background subtraction and variant clustering have led to new mathematical problems and solutions. Examples include quantification of allele fractions for both common and rare alleles in naturally mixed or intentionally pooled samples, detection of subtle target peaks using the deviation of the melting curve from exponential background, optimal mixing for genotyping nearest-neighbor-symmetric SNPs of various ploidy, and geometric combinatorial methods for thermodynamic parameter estimation.


Back to Data Analysis: qPCR BioStatistics & BioInformatics

Existing Methods and New Developmentsf or RT-PCR Analysis in R

Sebastian Kaiser1, Swanhild U Meyer2, Michael W Pfaffl2
1 LMU München, Germany; 2 TUM München, Germany

Abstract

Over the last years the software R made a strong impact on the research in all application fields dealing with statistical analysis.In bioscience the bioconductor packages include a wide range of tools for state of the art analysis of microarray data.All steps from data import, normalization, standardization, quality assessment, pattern recognition, assessment of significance of the findings, visualization, data output and much more are covered.This talk shows how most of this methods can be adopted for RT-qPCR analysis and additional methods for all areas of the workflow can be implemented.As an application example data from a multiplex RT-qPCR platform (TLDA) will be analysed. Additionally the adaption of two of the latest normalization techniques GPA and Mloess is demonstrated.


Back to Data Analysis: qPCR BioStatistics & BioInformatics

Pros and cons of single curve efficiency algorithms

Stefaan Derveaux1,2, Jo Vandesompele1,2, Jan Hellemans1,2
1 Center for Medical Genetics Ghent, Ghent University Hospital, Belgium; 2 Biogazelle, Ghent, Belgium

Abstract

The PCR amplification efficiency plays a critical role for accurate and reliable qPCR quantification. Serial dilution series of a mixture of samples representative of the whole set of samples in the experiment are currently considered as the gold standard to determine the efficiency of an assay. The use of such a standard curve requires a substantial amount of time and resources and relies on the assumption that all samples are amplified with the same efficiency, which may not always be true. To address these bottlenecks, various algorithms have been proposed to determine the efficiency of an individual qPCR reaction based on the raw fluorescent data curve. In this study, we evaluated two of those single curve efficiency determination algorithms, LinRegPCR and PCR Miner, and developed a framework to compare their accuracy and precision with the gold standard used serial dilution series. By using uniformity plates (plates with hundreds of replicates of the same reaction) we were able to determine the precision of the algorithms and by using different ratios of primers and competimers (P/C; C= non-functional primers that compete with the functional primers) in a reaction, we were able to modulate the amplification efficiency in a controlled manner. Both algorithms are often not accurate in that calculated efficiencies deviate from the gold standard. While the difference in efficiency varied between assays, both algorithms measured the same (relative) difference in efficiency between reactions with different P/C ratios as determined by the gold standard. The precision is relatively low, but similar to the one of a standard curve (obtained by iteratively calculating the efficiency using only one randomly selected replicate per dilution). As such, the single curve efficiency algorithms are possibly able to identify bad reactions which amplification efficiencies that greatly differ from that of the average sample population. This has great potential for sample and run quality control.


Back to Data Analysis: qPCR BioStatistics & BioInformatics

Bias in the Cq value observed with hydrolysis probe based quantitative PCR can be corrected with the estimated PCR efficiency value

Jan M Ruijter1, Jari M Tuomi2
1 Heart Failure Research Center, Amsterdam, the Netherlands; 2 Department of Physiology & Pharmacology, London, Canada

Abstract

For real-time monitoring of PCR amplification of DNA, quantitative PCR (qPCR) assays use various fluorescent reporters. DNA binding molecules and hybridization reporters (primers and probes) only fluoresce when bound to DNA and result in the non-cumulative increase in observed fluorescence. Hydrolysis reporters (TaqMan® probes and QZyme™ primers) become fluorescent during DNA elongation and the released fluorophore remains fluorescent during further cycles; this results in a cumulative increase in observed fluorescence. Although the quantification threshold is reached at a lower number of cycles when fluorescence accumulates, in qPCR analysis no distinction is made between the two types of data sets. Mathematical modeling shows that ignoring the cumulative nature of the data leaves the estimated PCR efficiency practically unaffected but will lead to at least 1 cycle underestimation of the quantification cycle (Cq value), corresponding to a 2-fold overestimation of target quantity. The effect on the target-reference ratio depends on the PCR efficiency of the target and reference amplicons. The leftward shift of the Cq value is dependent on the PCR efficiency and with sufficiently large Cq values, this shift is constant. This allows the Cq to be corrected and unbiased target quantities to be obtained.


Back to Data Analysis: qPCR BioStatistics & BioInformatics

Expression profiling – clusters of possibilities

Mikael Kubista1,2
1TATAA Biocenter; 2Institute of Biotechnology, Czech Academy of Sciences

Abstract

Advances in qPCR technology allow studies of increasingly large systems comprising many genes and samples. The increasing data sizes allow expression profiling both in the gene and the samples dimension, while also putting higher demands on sound statistical analysis and expertise to handle and interpret results. In this talk I present experimental design strategies for large studies, including proper design of multiplate experiments. I also present adequate pre-processing of primary qPCR data to prepare the measurement results for profiling analysis. Then I exemplify how powerful unsupervised expression profiling methods, including hierarchical clustering, heatmap, principal component analysis, and self-organizing maps, can be used to classify the expression of genes during the development of the frog Xenopus laevis and how the developmental stages from egg to tadpole can be clustered based on the genes’ expressions. Finally, I exemplify how powerful supervised methods, including potential curves, artificial neural networks, and support vector machines can be used to classify disease samples by comparing measured profiles of test samples with profiles measured on a reference set of samples.


Download a free GenEx trial version on GenEx.gene-quantification.info



Back to Data Analysis: qPCR BioStatistics & BioInformatics

Quality control for quantitative PCR based on amplification compatibility test

Ales Tichopad
Technical University Munich Germany

Abstract

Quantitative qPCR is a routinely used method for the accurate quantification of nucleic acids. Yet it may generate erroneous results if the amplification process is obscured by inhibition or generation of aberrant side-products such as primer dimers. Several methods have been established to control for pre-processing performance that rely on the introduction of a co-amplified reference sequence, however there is currently no method to allow for reliable control of the amplification process without directly modifying the sample mix. Herein we present a statistical approach based on multivariate analysis of the amplification response data generated in the real-time. The amplification trajectory in its most resolved and dynamic phase is fitted with a suitable model. Two parameters of this model, related to amplification efficiency, are then used for calculation of the Z-score statistics. Each studied sample is compared to a predefined reference set of reactions, typically calibration reactions. A probabilistic decision for each individual Z-score is then used to identify the majority of inhibited reactions in our experiments. We compare this approach to univariate methods using only the sample specific amplification efficiency as reporter of the compatibility. We demonstrate improved identification performance using the multivariate approach compared to the univariate approach. Finally we stress that the performance of the amplification compatibility test as a quality control procedure depends on the quality of the reference set.


Back to Data Analysis: qPCR BioStatistics & BioInformatics