Fig. 2

Characterization and result of the functional analysis of Module 1 (turquoise). a M1 is correlated with 5 of the 7 continuous phenotypes, including Novel Object (p > 6E-04, r = 0.72), Novel Place (p > 1E-02, r = 0.57) as well as mBDNF (p > 3E-03, r = 0.65) GLT1 (p > 1E-02, r = 0.56) and pSer31Th (p > 1E-04, r = 0.76). b Heatmap rows correspond to genes and columns to samples, where red denotes up-regulation of gene expression and green denotes down-regulation. The corresponding graph shows corresponding Module EigenGene (ME) expression values across the samples grouped by experimental group assignment. The ME represents the overall gene expression profile of the module, and has positive value for an array when a lot of the module genes are up-regulated (red in the heatmap) and has negative values when a lot of the module genes are down-regulated (green in the heatmap). The genes are largely down-regulated in WT-Veh and ASO-Veh, and up-regulated in WT-NSC and ASO-NSC. c Cell-type specific expression analysis (CSEA) identified candidate cell populations associated to M1. Each cell type is represented by a multilayer hexagonal shape in a single bullseye plot [108]. The size of hexagons for each cell type is scaled to the number of specific and enriched transcripts at four different stringency thresholds (pSI: 0.05, 0.01, 0.001, 0.0001). The more centered hexagon the more stringent threshold is and color coded by Fisher’s exact test p values as shown [108]. M1 is enriched in genes expressed in oligodendrocyte, spiny GABAergic neurons and less significantly in astrocytes. For example, striatal D1+ spiny neurons have many unique transcripts (large fourth hexagon at pSI < 0.05), while cortical oligodendrocytes have many unique transcripts at several stringency level (pSI < 0.05, pSI < 0.01, pSI < 0.001). d Representative functional annotation terms of M1 (all functional annotations can be found in Additional files 7, 8, 9 and 10)