Objective
The Combined Analysis of Psychiatric Studies (CAPS) project conducted extensive review and regularization across studies of all schizophrenia linkage data available as of 2011 from the NIMH-funded Center for Collaborative Genomic Studies on Mental Disorders, also known as the Human Genetics Initiative (HGI). Here we reanalyze the data using statistical methods tailored to accumulation of evidence across multiple, potentially highly heterogeneous, sets of data.
Method
Data were subdivided based on contributing study, major population group, and presence or absence within families of schizophrenia with a substantial affective component. The Posterior Probability of Linkage (PPL) statistical framework was used to sequentially update linkage evidence across these data subsets (omnibus results).
Results
While some loci previously implicated using the HGI data were also identified in our omnibus analysis (2q36.1, 15q23), others were not. Several loci were found that had not been previously reported in the HGI samples, but which are supported by independent linkage and/or association studies (3q28, 12q23.1, 11p11.2, Xq26.1). Not surprisingly, differences were seen across population groups. Of particular interest are signals on 11p15.3, 11p11.2, and Xq26.1, for which families with substantial affective component support linkage while the remaining families give evidence against linkage. All three of these loci overlap with loci reported in independent studies of bipolar disorder or mixed bipolar-schizophrenia samples.
Conclusions
Public data repositories provide the opportunity to leverage large, multi-site data sets for studying complex disorders. Analysis with a statistical method specifically designed for such data enables us to extract new information from an existing data resource.