Back

Clustering schizophrenia genes by their temporal expression patterns aids functional interpretation: genetics-based evidence in favor of the two-hit hypothesis

van der Meer, D.; Cheng, W.; Rokicki, J.; Fernandez-Cabello, S.; Shadrin, A.; Smeland, O. B.; Ehrhart, F.; Guloksuz, S.; Steen, N. E.; Djurovic, S.; Westlye, L. T.; Andreassen, O. A.; Kaufmann, T.

2022-08-25 psychiatry and clinical psychology
10.1101/2022.08.25.22279215 medRxiv
Show abstract

Schizophrenia is a highly heritable brain disorder with a typical symptom onset in early adulthood. The two-hit hypothesis posits that schizophrenia results from deviant early neurodevelopment, predisposing an individual, followed by a disruption of later brain maturational processes that trigger the onset of symptoms. Here, we investigate how the timing of expression of 345 putative schizophrenia risk genes may aid in understanding the interplay of neurobiological processes in the pathophysiology of schizophrenia. Clustering of brain transcriptomic data across the lifespan revealed a set of 183 genes that was significantly upregulated prenatally and downregulated postnatally and 162 genes that showed the opposite pattern. The prenatally upregulated set of genes was functionally annotated to fundamental cell cycle processes, while the postnatally upregulated set was associated with the immune system and neuronal communication. We subsequently calculated two set-specific polygenic risk scores for 743 individuals with schizophrenia and 743 sex- and age-matched healthy controls. We found an interaction between the two scores; higher prenatal polygenic risk was only significantly associated with schizophrenia diagnosis and severity, at higher levels of postnatal polygenic risk. We therefore provide genetics-based evidence in favor of the two-hit hypothesis, supporting that schizophrenia may be shaped by disruptions of separable biological processes acting at distinct phases of neurodevelopment.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.