The gene expression profiles were filtered to select genes with a reliable expression change by applying the criterion of a two-fold change (in the same direction) in expression level in two or more consecutive time points.
Anesthesia was considered successful if the subject did not respond to the maximum output of the pulp tester at two or more consecutive time points.
Next, we calculated elongation rates for genes that had two or more consecutive transition points.
We selected genes that were significantly regulated (p < 0.01) at two or more consecutive time points or in replicate measurements, which yielded 1508 LPA-regulated genes (see additional file 1: complete dataset).
In order to reduce the likelihood of false-positives due to a large number of t-tests, we considered MMNs to be significantly different when the P-values were lower than 0.05 at 10 (= 20 ms) or more consecutive time points (Guthrie and Buchwald 1991).
For this, we selected probes up- or down- regulated more than 3 or 1.5-fold in two or more consecutive time-points, with the additional constraint that the expression level of the selected probes should not vary over 1.5-fold outside these consecutive time-points (Fig. S3).
Out of the 113 published genes, none met our filtering criteria of 1.5-fold change in two or more consecutive time-points.
Using more conservative filtering criteria (3 fold change in two or more consecutive time-points) we were unable to detect any transcriptional alterations in either endoneurial or neuronal compartments of diabetic mice.
