A specific question we are examining is whether risk for typically adult-onset mental illness relates to traits in our youth clinical cohort that could be amenable to intervention. We have found an association between polygenic risk for SCZ and working memory across diagnoses in our pilot sample.

We have shown that executive cognition associates with dimensions of autism spectrum disorder (ASD), bipolar disorder and psychosis that can be modeled as a general psychopathology factor. We are now extending this model to include common genetic variation. We are also conducting randomized, double-blind placebo-controlled trials of computerized cognitive training in youth with neuropsychiatric illness.

Our work suggests that the developmental timing of exposure to adversity is an important determinant of depression.

Memory, and other executive functioning deficits, are common among people exposed to adversity.

We have also developed a novel metabolomics method that can measure >200 metabolites simultaneously in a small biospecimen volume. We have identified several key metabolites associated with outcome, including the endocannabinoid anandamide, which may serve as a prognostic biomarker for secondary brain injury. In other studies, we have identified that impaired oxidative phosphorylation may be linked to atrial fibrillation-associated stroke risk. These studies may help guide secondary stroke risk reduction, particularly in patients who have embolic stroke of undetermined source.

Cross Disorder Analysis of Five Neuropsychiatric Disorders. In the latest cross disorder analysis study published at Lancet (Smoller, Ripke, Lee, et al. 2013), we aimed to identify specific genetic risk variants shared between the five disorders in the Psychiatric Genomics Consortium: autism spectrum disorder, attention deficit-hyperactivity disorder, bipolar disorder, major depressive disorder, and schizophrenia. Using genome-wide single-nucleotide polymorphism (SNP) data for the five disorders in 33,332 cases and 27,888 controls, I applied a multinomial logistic regression procedure with model selection (Lee et al. 2011) to identify the best-fitting model of relations between genotype and phenotype. Model selection analysis supported effects of these loci for several disorders. Loci previously associated with bipolar disorder or schizophrenia had variable diagnostic specificity. I also performed pathway analyses using INRICH (Lee et al. 2012) to establish the biological associations underlying genetic overlap for the five disorders. This multivariate analysis supported a role for calcium channel signaling genes for all five disorders. Finally, I used enrichment analysis of expression quantitative trait loci (eQTL) data and showed that SNPs with evidence of cross-disorder association were enriched for brain eQTL markers.

Partitioning Heritability Analysis to Study the Complex Neurogenetic Architecture of Schizophrenia. Schizophrenia is a devastating neurodevelopmental disorder with a complex genetic etiology. Widespread cortical gray matter loss has been observed in patients and prodromal samples. However, it remains unresolved whether schizophrenia-associated cortical structure variations arise due to disease etiology or secondary to the illness. Our recent work published in Molecular Psychiatry (Lee et al. 2016) addressed this question using a partitioning-based heritability analysis of genome-wide single-nucleotide polymorphism (SNP) and neuroimaging data from 1,750 healthy individuals. We find that schizophrenia-associated genetic variants explain a significantly enriched proportion of trait heritability in eight brain phenotypes (false discovery rate=10%). In particular, intracranial volume and left superior frontal gyrus thickness exhibit significant and robust associations with schizophrenia genetic risk under varying SNP selection conditions. Cross-disorder comparison suggests that the neurogenetic architecture of schizophrenia-associated brain regions is, at least in part, shared with other psychiatric disorders. Our study highlighted key neuroanatomical correlates of schizophrenia genetic risk in the general population. These may provide fundamental insights into the complex pathophysiology of the illness, and a potential link to neurocognitive deficits shaping the disorder.

Huntington’s disease motor onset modifier genes located. Top locus does not modify disease duration. 

Huntington’s disease CAG repeat mutation elicits effects from conception  

Leveraging (1) advanced MRI measurements of physiologic parameters to probe white matter microvascular biology; (2) genetic manipulation of human in-vitro models of BBB and; (3) functional in-vitro assays we unravel the molecular mechanisms by which ACBD1 increases the permeability to inflammatory cells. This combined in-vivo and in-vitro approach challenged current paradigms in inflammatory and degenerative disorders of the white matter by demonstrating that we can test the effect of single genes upon BBB function in humans, in a safe and feasible manner to answer clinically relevant, biological questions.

We characterized the relative effects of genetic variation and dietary and intravenous sodium challenge on ANP production in translational work in the MGH Clinical Research Center.

We developed one of the largest cellular biobanks in the world for the study of brain diseases, a key resource for understanding how disease-associated genes contribute to risk, as well as screening for new treatments.

We developed machine learning approaches that apply electronic health records to predict suicide and other causes of mortality in individuals with psychiatric illness.

Our results suggest that both circadian rhythm and sleep traits share genes and pathways with psychiatric diseases such as schizophrenia, depression and autism, and metabolic diseases such as obesity and diabetes.

Leveraging longitudinal electronic health records (EHRs) for phenotyping, genomic analysis and risk prediction. We have developed algorithms enabling high-throughput phenotyping for large-scale genomic studies using natural language processing of EHRs. We also developed and validated EHR-based models that were able to identify nearly half of all suicides and suicidal behaviors an average of 3 – 4 years in advance.

Figure 2. We examined a cohort of 238 patients with SMA with 1,130 creatinine (Crn) observations. Decreased Crn levels reflected disease severity, suggesting that Crn is a candidate biomarker for SMA progression. We conclude that Crn measurements should be included in the routine analysis of all patients with SMA. In future studies, it will be important to determine whether Crn levels respond to molecular and gene therapies.

Reference: Alves et al., Neurology, 2020; 94:1-11. DOI: 10.1212/WNL.0000000000008762

Familial hypercholesterolemia mutations are present in 2% of those with severely elevated cholesterol levels, but – if present – confer markedly increased risk of coronary artery disease.

A genome-wide polygenic score for coronary artery disease identifies 8% of the population with risk equivalent to a rare monogenic mutation.