Schizophrenia (SCZ) is a severe psychiatric disorder with a genetic susceptibility. Alterations in neurochemical signaling, as well as changes in brain structure and function, manifest during the course of SCZ and are likely causative of the symptoms shown by affected individuals. However, little is known about the timing of these changes, particularly in the pre-morbid and prodromal phases of SCZ. Here, we performed a gene-based and pathway-based meta-analysis of 5 microarray datasets from human induced pluripotent stem cells (hiPSCs)-derived neurons and post-mortem brain tissue from SCZ and healthy controls (HC), with the underlying assumption they might represent the neurobiological make-up of SCZ in the pre-morbid and chronic stages of illness, respectively. Thus, we identified 1 microarray expression profiling dataset of hiPSCs-derived neurons (GSE25673) and performed a systematic search of microarray expression profiling datasets from SCZ post-mortem brain publicly available on the Gene Expression Omnibus (GEO) repository. We selected 4 different SCZ post-mortem brain microarray expression profiling datasets (GSE17612, GSE21935, GSE12649, and GSE21338) according to specific inclusion and exclusion criteria. We downloaded raw data and performed quality controls, differential expression analysis, and gene-based, as well as pathway-based meta-analysis. Neuronal pentraxin 2 (NPTX2) gene was consistently down-regulated across all datasets, with highly significant association in the meta-analysis (FDR<1.0E-04). These results highlight the heuristic value of microarray meta-analysis and suggest a role of NPTX2 as a disease biomarker, provided that it achieves biological validation in future studies examining whether this down-regulation has predictive value with respect to the developmental trajectory of SCZ.
Keywords: Biostatistics; Human induced pluripotent stem cells; Meta-analysis; Pathway analysis; Staging; Transcriptomics.
Copyright © 2017 Elsevier B.V. All rights reserved.