We used DESeq2 [52] to determine differential expression from the corrected RNA-seq read count values, analyzing only transcripts with an average FPKM ≥ 1 across all samples. Considering that neuropsychiatric disorders, including SZ, are highly heterogeneous, we determined statistically significant differences in gene expression between SZ samples and controls at a nominal p value (p < 0.05), an approach similarly taken in many previous SZ transcriptomic studies [25, 53–58], but we also applied a multiple comparison correction to the p-values to compute for false discovery rate (FDR) [59].
