Fig. 7: Expression of early neurogenic programs in neurodevelopmental cell-types from human fetal cortex. | Nature Communications

Fig. 7: Expression of early neurogenic programs in neurodevelopmental cell-types from human fetal cortex.

From: Transcriptional programs regulating neuronal differentiation are disrupted in DLG2 knockout human embryonic stem cells and enriched for schizophrenia and related disorders risk variants

Fig. 7: Expression of early neurogenic programs in neurodevelopmental cell-types from human fetal cortex.

a Identification of programs enriched/depleted for genes with high expression in CA1 pyramidal neurons relative to other brain cell-types (pyramidalhigh); enrichment compared to all expressed genes (two-sided Fisher’s Exact Test). Raw and Bonferroni-corrected p values given (Poverlap, Poverlapcorrected), correction is for the 4 tests performed. b Pyramidalhigh genes were partitioned based on their overlap with neurogenic programs. Each segment of the Venn diagram shows the number of genes in each subset and the regression coefficient (β) and uncorrected p value (P) for SZ common variant enrichment (one-sided MAGMA test), conditioning on all expressed genes. Bold indicates enrichments surviving correction for 7 tests. c A one-sided Fisher’s Exact Test was used to identify key regulators whose known/predicted targets are over-represented amongst pyramidalhigh genes overlapping the late neurogenic program when compared to all expressed genes; both raw and Bonferroni-corrected p values are given (Poverlap, Poverlapcorrected), where correction is for the 7 regulator sets tested. d Cells corresponding to distinct neurodevelopmental cell-types (RG radial glia; IPC Intermediate progenitor cell), including cell-types at different stages of maturity, were identified and extracted from a previously published single-cell RNA-seq study of human fetal cortex across peak stages of neurogenesis32 (Methods). >80% of genes belonging to each in vitro-defined early neurogenic program and 55% of the late-pyramidalhigh set (top RHS) were present in the fetal data (early-transient−/− n = 665 genes; early-stable−/− n = 1484 genes; early-increasing−/− n = 431 genes; late-pyramidalhigh n = 101 genes). For each program, the mean and standard error of gene-level expression averages (see main text/Methods) were calculated for each fetal cell-type/developmental stage. The fetal data captures both direct (RG ➝ neuron) and indirect (RG ➝ IPC ➝ neuron) neurogenic pathways. The (deep layer) neurons present in our day 30–60 cultures in vitro are predominantly born via the direct neurogenic pathway. For each program, differences between gene-level averages for successive cell-types/stages were evaluated using a two-tailed Student’s t test and p values Bonferroni-corrected for 6 pairwise comparisons. *p < 0.05; **p < 0.01; ****p < 0.0001. All data presented as mean ± SEM. Bold indicates tests surviving Bonferroni correction. The exact p values are provided in Supplementary Data 10.

Back to article page