Immune signatures and disorder-specific

surpassing that attributed to more physical disorders. Worldwide prevalence rates for the five major psychiatric disorders range from 0.2% for attention-deficit hyperactivity disorder (ADHD) to 5.7% for major depressive disorder (MDD), with schizophrenia, autism spectrum disorder (ASD) and bipolar disorder showing intermediate prevalence. Although these disorders are classified into distinct disorder categories, they show overlap in symptomatology and shared genetic risk in family studies, as well as comorbidity in individuals and families. For example, 41% of patients with ASD were found to have a comorbid disorder, including ADHD. In addition, family studies show that relatives of probands with bipolar disorder show an increased risk for both bipolar disorder and schizophrenia. This could indicate common aetiological pathways for some psychiatric disorders and studying them jointly could lead to new insights. Twin study heritability estimates for psychiatric disorders are moderate for MDD (~40%) and high (80–90%) for ADHD, ASD and schizophrenia, indicating a significant genetic component in their aetiology. Large genome-wide studies have now started to reveal risk variants for individual disorders but an outstanding question is whether genetic risk factors or polygenic risk scores are specific to particular disorders, or shared. For instance, certain copy number variations (CNVs) are overrepresented in both patients with schizophrenia and those with ASD. Additionally, calcium-channel activity genes have been shown to have shared involvement in ASD, schizophrenia and bipolar disorder, whereas polygenic risk analysis showed strong overlap of genome-wide risk between bipolar disorder, MDD and schizophrenia, and moderate overlap between ASD and schizophrenia. In another sample, high sharing was seen between bipolar disorder and schizophrenia, with moderate overlap between MDD and schizophrenia, bipolar disorder and ADHD and low, but significant, overlap between schizophrenia and ASD. These results suggest that some shared molecular genetic factors underlie a significant proportion of the risk for development of several psychiatric disorders. To date, however, it is not known whether these genomic overlaps are reflected in overlapping patterns of gene expression in tissues. In this study we therefore explored whole blood gene expression across adult and childhood ADHD, ASD, MDD and healthy controls, using weighted gene co-expression analysis to search for patterns of correlated gene expression between disorders as well as disorder-specific gene expression signatures. In addition, we generate polygenic risk scores to assess overlap between disorders on a genome-wide genetic-risk level. We next investigate whether differences in polygenic risk scores are reflected in disorder-related gene expression profiles and whether they can be used to tease apart genetic and environmental influences on gene expression.