Copy number and single nucleotide variants in neuropsychiatric diseases: a cross-disorder whole-genome sequencing study

Aims Recent studies have provided evidence of a substantial genetic contribution to risk for neuropsychiatric diseases including bipolar disorder, major depression and schizophrenia. In this work, we performed whole-genome sequencing (WGS) for individuals with multiple neuropsychiatric diseases as well as matched healthy controls to identify genomic variations including single nucleotide variants (SNVs), copy number (CNVs), and structural variations (SVs). Methods A cohort of 95 individuals was studied, spanning the following DSM-IV-defined disorders: bipolar disorder (25 subjects), major depressive disorder (14 subjects), schizoaffective disorder (6 subjects), schizophrenia (21 subjects) and healthy controls (29 subjects). SNVs were called using HaplotypeCaller, and the variants with minor allele frequency (MAF) > 1% in three cohorts (1000G, ExAC, and ESP6500) and those identified in healthy controls were filtered out. CNVs were identified using a recent method developed by our group called BIC-seq2. Results We identified a total of 19 million variants (85%) that passed the variant quality score recalibration step, with an average of 3.2~4.0 million variants per individual. No significant differences between disorders were observed in terms of number of variants. Most of the variants were intergenic SNVs (57.5%) and intronic SNVs (33.8%). After filtering common and non-causal variants, each individual had, on average, 55,780 variants. A total of 367 regions showing a significant copy number gain (72) or loss (295) in at least two individuals with the same disorder were identified. 52 recurrent CNV calls overlap exonic regions and include 616 overlapping SNVs. Interestingly, many genes associated to these exonic regions were linked with fetal development and prenatal growth, processes known to have a high impact on neuropsychiatric diseases.