Genetics of eeg oscillations reveal novel biological insights into the links between brain structure, brain function, and behavior

Oscillations in neuronal brain activity play a crucial role in information processing and have extensively been studied as biological markers of human behavior and psychopathology. Although traditional twin studies suggest that brain oscillations are highly heritable, our understanding of their molecular genetic architecture is still very limited. Here we conducted a genome-wide association study (GWAS) of resting-state EEG oscillations to discover associated genomic loci and to examine the pleiotropic relationships with other complex traits, i.e., the links with brain structure and mental illness. We present findings from the largest resting-state EEG brainwave GWAS to date, which combined data from nine cohorts and included up to 14,361 participants. Using harmonized analysis protocols, we examined the power of the EEG frequency bands alpha, beta, delta, theta, and broadband at the vertex site, as well as alpha power and alpha peak frequency at occipital leads. GWAS analyses were run in RAREMETALWORKER followed by meta-analysis in METAL. We used bioinformatic tools such as LD score regression and pleioFDR to investigate a shared genetic basis with other complex traits. Heritability estimates (h2SNP) ranged from 14% to 27% (SE: 3.7%). We identified two genome-wide significant loci, i.e., an intergenic region at 13q12.3 (p=6.6e-09), and an intronic region of FANCA at 16q24.3 (p=1.4E-08), both associated with alpha peak frequency. We discovered 32 additional genomic loci by leveraging pleiotropy with psychiatric traits such as schizophrenia, major depression, and bipolar disorder. Using genetic correlations, we demonstrate a shared genetic basis between EEG power and MRI-derived cerebral white matter volume (rG=-0.33, p=3.0e-07) as well as cortical surface area (rG=-0.26, p=2.0e-04), with the top regional associations implicating the orbitofrontal, anterior cingulate, and precuneus surface area. Genetic correlations also indicated a genetic overlap with generalized epilepsy and depressive-like symptoms. Our results suggest that common genetic variation substantially affects EEG oscillations, with a genetic architecture that overlaps those of brain structure variables. Our results provide support for a pleiotropy with psychiatric and neurological traits. In sum, we corroborate previous findings on the heritability of brain oscillations and provide novel biological insights into the links between brain structure, brain function, and behavior.