Gene expression analysis of peripheral blood leukocytes from discordant sib-pairs with schizophrenia and bipolar disorder reveals points of convergence between genetic and functional genomic approaches

We performed global RNA transcript analysis and comprehensive gene group analysis of peripheral blood leukocyte (PBL) RNA from two groups of matched sib-pairs that were discordant for either schizophrenia (n = 33 sib-pairs) or bipolar disorder (n = 5 sib-pairs). The pairs chosen for these analyses were selected from families with known patterns of genetic linkage (5q for schizophrenia and 6q for bipolar disorder). At the single gene level, we obtained lists of the transcripts with the most significant changes in expression and from these lists determined those with the highest degree of predictive power for classifying subjects according to diagnosis in these samples. At the gene group level, we comprehensively analyzed pairwise expression changes of more than 4,000 functional groups and cytogenetic locations, and present a novel method of displaying these data that we term "cytogenomic" mapping. Verification of selected changes in expression was performed using quantitative real-time RT-PCR. Our results provide compelling evidence for the utility of analyzing PBL RNA for changes in expression in neuropsychiatric disorders.