PROJECT SUMMARY
Neurocognitive deficits represent a core component of several major neuropsychiatric disorders, including
schizophrenia, affective disorders, autism spectrum disorders, and attention-deficit/hyperactivity disorder, and
are not merely an epiphenomenon reflecting symptom severity, environmental deprivation, or medication side
effects. The centrality of cognition to mental health is reflected in the RDoC matrix, in which “Cognitive
Systems” is one of the six fundamental domains of investigation. Moreover, cognitive deficits are associated
with poor functional outcomes, and are generally not well treated by existing psychotropic medications. Based
on family studies and longitudinal observations, it has long been recognized that cognitive abnormalities are (to
a significant degree) genetically mediated endophenotypes of serious mental illnesses. The Cognitive
Genomics Consortium (COGENT), led by the PI of this application, has conducted a series of genome-wide
association studies (GWAS) of cognitive phenotypes, which have confirmed the significant genetic overlap
between cognitive performance and most forms of psychiatric illness at the molecular level.
During five years of NIMH funding, this project has resulted in 13 publications, including the largest cognitive
GWAS to date (N=373,617 yielding 241 significant loci) and the first large exome study identifying rare variants
associated with cognitive performance in the general population. Moreover, we applied novel approaches to
dissecting the pleiotropy observed between cognition and psychopathology, identifying “meta-loci” which are
regions of the genome that underly specific patterns of pleiotropic overlap between phenotypes. This work has
yielded new treatment targets, as well as biological insights on a dissociation between neurodevelopmental
(prenatal) pathways vs adult synaptic processes underlying distinct forms of cognitive function and dysfunction.
Over the next five years, we propose three main aims to extend our previous work, with an emphasis on
biological interpretability and potential clinical applicability of results. First, we will add new, ethnically diverse
cohorts for further cognitive GWAS and rare variant analysis, and we will seek to identify key subdomains of
cognitive function for further downstream analysis. Second, we will incorporate these cognitive subdomains as
well as newly available, large-scale neuroimaging genomics data in a novel set of analyses, derived from our
“meta-loci” approach, to parse psychiatric phenotypes into biologically coherent subcomponents. Third, we will
construct novel, biologically-informed polygenic risk scores (PRS) and test whether these have greater
prognostic value as compared to conventional, genome-wide PRS based on single-trait GWAS. Thus, our plan
leverages diverse large-scale genomics resources, and a range of expertise, to derive actionable information
(novel treatment targets, biological mechanisms, and biomarkers) relating to a range of psychiatric disorders.