Relative quantification data analysis and software applications

A major challenge is the development of exact and reliable gene expression analysis and quantification software. A 'one-fits-all' detection and application software is the target for future developments and seems the optimal solution. But can we implement various detection chemistries with varying background and fluorescence acquisition modes in one software package? Should we not think about optimized models on each real-time platform and for each applied chemistry? In biological research and in clinical diagnostics, real-time qRT-PCR is the method of choice for expression profiling. On the one hand cycler and chemistry developed much faster than detection and analysis software. However, accurate and straightforward mathematical and statistical analysis of the raw data (cycle threshold/crossing point values or molecules quantified) as well as the management of growing data sets have become the major hurdles in gene expression analyses. Now the 96-well applications are the standard in the research laboratories, but in the near future high throughput 384-well applications will generate huge amounts of data. The data need to be grouped (Hellemans et al., 2006) and standardized by intelligent algorithms. Real-time qPCR data should be analyzed according to automated statistical method, e.g. Kinetic Outlier Detection (KOD), to detect samples with dissimilar efficiencies (Bar et al., 2003). Mostly the statistical data analysis or CP values is performed on the basis of classical standard parametric tests, such as analysis of variance or t-tests. Parametric tests depend on assumptions, such as normality of distributions, whose validity is unclear (Sheskin, 2000). In absolute or relative quantification analysis, where the quantities of interest are derived from ratios and variances can be high, normal distributions might not be expected, and it is unclear how a parametric test could best be constructed (Pfaffl et al., 2002; Sheskin, 2000). At present, the following relative quantification data analysis and software applications are available.

0 0

Post a comment