T192. Weighted gene coexpression network analysis of IPSC generated from patients with schizophrenia

Abstract

Background. Human induced pluripotent stem cells (hiPSC) have provided a new way of studying Schizophrenia (SZ), by allowing the establishment of cellular models accounting for the patient genetic background. Here we conducted an exploratory RNA-sequencing profiling study of different cell lines derived from hiPSCs generated from somatic cells of subjects from the population isolate of the Central Valley of Costa Rica (CVCR). Methods. Lymphoblastoid cells lines (LCLs) were transformed into hiPSCs, Neuronal Precursors Cells (NPC) and cortical neurons, using well establish methodology. RNA from these cells were then sequenced using Illumina HiSeqTM2500. Raw count data measured 48162 transcripts across all samples. The 15000 more expressed genes were subjected to VOOM normalization in EdgeR, a variance-stabilization transformation method. Normalized values were used as input for weighted gene coexpression network analysis (WGCNA). Differential expression of MEs (module eigengene) comparing healthy controls and patients with schizophrenia across all cell types were performed. Results. On total 4 cell lines (LCL, hiPSC, NPC and cortical neurons) of 6 healthy controls (HC) and 7 SZ patients from the CVCR were included on the WGCNA analysis. Biweight midcorrelation was used to define the coexpression similarity, resulting in 129 modules. Differential expression of MEs were observed on relation to phenotype (p=0.04) and presence of NRG1 Val66p.Leu mutation (p=0.02). Noteworthy, MEs were able to separate members of the same family from other subjects (p=0.0006). Conclusions. Our study used WGCNA to establish blocks of gene expression on a hiPSC cellular model of SZ related to phenotype and genotype.