Moreover, seven patients (33.3%) experienced a complete biological response with a normalisation of PSA following treatment.
Among the 36 patients with at least two assessments of serum AFP, one complete and four partial biological responses (defined by normalisation and decrease level more than 50%, respectively) were observed.
For all 331 microarrays with available CEL files (and carried out for biological replicates), GC-RMA normalisation was carried out.
However, despite a relatively large sample size, when inappropriate reference genes were used for normalisation, a true biological difference in expression between tumour and normal was not detected.
The mean μ is proportional to the concentration of the transcript of interest in the original biological sample, up to a normalisation factor specific to the lane of the sequencer.
The diagnostic eight-pack array consists of eight mini arrays with 1900 features each, and these features comprise 255 positive controls, 100 negative controls and 1545 biological sequences used for data normalisation and sample classification.
In the first step of the statistical analysis the arrays were normalised by the invariant set normalisation method, so that all non-biological variables were reduced, followed by a model-based expression analysis, using the model PM-only.
Complete biological response was defined as a complete normalisation of all tumour markers and partial biological response was defined as a decrease of more than 50% compared to baseline level.
Other experiments at our facility using the more recent Illumina Human HT-12 and Mouse Ref 8 BeadChips exhibit similar batch effects to those in this study; samples are observed to cluster preferentially with others processed in the same run, rather than by the biological differences between them, even after quantile normalisation (data not shown).
However, accurate and reliable interpretation of RT-qPCR results depends heavily on the use of suitable reference genes for normalisation to eliminate or minimise non-biological variation between test samples.
Normalisation was carried out by Robust Multichip Analysis (RMA) and differentially expressed contigs were identified across the normalised microarray datasets for biological replicates using linear modelling in limma in R v2.7.1 (Linear Models for Microarray Data) [57].
