miR-34a, two controls and one bipolar sample were excluded for very low normalizer expression for ANK3 and CACNB3 measurements. Student t-test for unpaired data with equal variance was applied for the Sholl analyses. In order to assess if there was evidence within the NanoString profiles for common molecular pathways being affected by miR-34a over expression, since the selection of 131 genes for inclusion in the NanoString probe set was purposely biased to cover genes implicated by neuropsychiatric genetics and ion channels, we deemed consideration of enrichment with Gene Ontology or related categories inappropriate. As an alternative computational approach to the systems-based modeling of the differential gene expression data, we determined if there were statistically significant properties of the protein-protein interaction network amongst the validated miR-34a target genes using VisANT 4.0 (http://visant.bu.edu) and/or the proteins encoded by the differentially expressed genes using DAPPLE v2.0 (Disease Association Protein-Protein Link Evaluator) (Figure 4).
