Genetic models of schizophrenia and related psychotic disorders: progress and pitfalls across the methodological "minefield".

The challenge of modelling a complex and multifaceted disorder such as schizophrenia is epitomised by the considerable degree of phenotypic variability described in patients and by the absence of specific and consistent neuropathological biomarkers. The pattern and severity of a range of clinical features, including florid psychotic symptoms such as hallucinations and delusions, negative symptoms and cognitive dysfunction, together with age at onset, course of illness and other indices, can vary greatly between individual patients. The undefined nature of the relationship between diagnosis and underlying aetiology has complicated research in the field of clinical and preclinical neuroscience, thereby making it difficult to generate or evaluate appropriate disease models of schizophrenia. In the present review, we explore those conceptual and practical issues that relate specifically to the genetic modelling of schizophrenia and related disorders in rodents. Practical issues that impact on the robustness of endophenotypic findings and their translational relevance are discussed with reference to evidence from selective genetic models of candidate risk genes and copy number variants implicated in schizophrenia.