The results of known targets and enriched regions show consistent differences between the various normalization approaches.
A potential cause of the observed difference between the tested normalization approaches is observed in the results on global level: the ranking of probes changes when using some normalization approaches, increasing the likelihood of un-enriched probes being spread over the whole dynamic range of the enriched probe distribution.
In our experiments, we compare different views from the speaker and pose normalization strategies learned and tested on the same data and the results, therefore, reflect differences between the views and pose normalization strategies rather than differences in the test datasets.
We can compare the AIC and ABIC values among the MLE based models and among the Bayesian models, respectively, although we cannot directly compare the AIC value with ABIC values here because we did not adjust the difference in the normalization factors between AIC and ABIC in the considered models.
For simulation type 2, the differences between normalization methods (excepted FPKM) are less important.
Based on the highly significant differences between groups from this study, we concluded that CSF normalization was able to detect differences between healthy and pathologic IVD whereas bone normalization detected more differences between the severities of these pathologies.
We use two different base counts for normalization, to reflect fundamental differences between the tools with respect to selective use of markers: the first count is the total number of markers present in the SBF as generated by the tool, denoted E(X), while the second is the total number of markers present in the genome within the coordinates of the generated blocks, denoted E(X′).
Due to scale differences between the arrays we also conducted global scale normalization across arrays.
In general, there are many differences between the MIDI data and user's humming data, and normalization is required.
