The polygenic risk for bipolar disorder influences brain regional function relating to visual and default state processing of emotional information

Abstract

Genome-wise association studies have identified a number of common single-nucleotide polymorphisms (SNPs), each of small effect, associated with risk to bipolar disorder (BD). Several risk-conferring SNPs have been individually shown to influence regional brain activation thus linking genetic risk for BD to altered brain function. The current study examined whether the polygenic risk score method, which models the cumulative load of all known risk-conferring SNPs, may be useful in the identification of brain regions whose function may be related to the polygenic architecture of BD. We calculated the individual polygenic risk score for BD (PGR-BD) in forty-one patients with the disorder, twenty-five unaffected first-degree relatives and forty-six unrelated healthy controls using the most recent Psychiatric Genomics Consortium data. Functional magnetic resonance imaging was used to define task-related brain activation patterns in response to facial affect and working memory processing. We found significant effects of the PGR-BD score on task-related activation irrespective of diagnostic group. There was a negative association between the PGR-BD score and activation in the visual association cortex during facial affect processing. In contrast, the PGR-BD score was associated with failure to deactivate the ventromedial prefrontal region of the default mode network during working memory processing. These results are consistent with the threshold-liability model of BD, and demonstrate the usefulness of the PGR-BD score in identifying brain functional alternations associated with vulnerability to BD. Additionally, our findings suggest that the polygenic architecture of BD is not regionally confined but impacts on the task-dependent recruitment of multiple brain regions.

Polygenic risk scores distinguish patients from non‐affected adult relatives and from normal controls in schizophrenia and bipolar disorder multi‐affected kindreds

“…Our study has strengths and limitations. A first strength is the combination of a relatively large sample size from multigenerational SZ and BD families and the utilization of a PRS built with the most recent results in term of association of risk alleles with SZ and BD (Hou et al, ; Schizophrenia Working Group of the Psychiatric Genomics, ) and with optimal numbers of risk alleles to compare patients to controls, which was not the case in previous studies (Dima et al ; Fullerton et al, ). Second, for the imputation process, we used family information which improves the phasing of the genotypes (O'Connell et al, ).…”

“…Using only a BD PRS, Fullerton et al () compared BD affected subjects and their NAARs, but only 32 SNPs were used in the PRS, the number of subjects per family was relatively small and the subjects had diverse ethnic backgrounds. Another study tried to identify brain regions whose function may be related to BD PRS and also examined the discrimination by the BD PRS of BD patients from unaffected first‐degree relative and unrelated healthy controls (Dima, de Jong, Breen, & Frangou, ), but the group sizes were small. A study of Ahn, An, Shugart, and Rapoport () also found that childhood‐onset SZ patients had higher genetic risk scores for SZ than their healthy siblings with a SZ PRS defined from the PGC results.…”

Telomere length as a predictor of emotional processing in the brain

“…Data quality was controlled in PLINK v1.07 (Purcell et al, 2007) using the same parameters as described in Coleman et al (2016) and Dima, de Jong, Breen, and Frangou (2016). Data quality was controlled in PLINK v1.07 (Purcell et al, 2007) using the same parameters as described in Coleman et al (2016) and Dima, de Jong, Breen, and Frangou (2016).…”

“…DNA extracted from buccal swabs was genotyped on the Psych Chip (Illumina Infinium PsychArray-24). Data quality was controlled in PLINK v1.07 (Purcell et al, 2007) using the same parameters as described in Coleman et al (2016) and Dima, de Jong, Breen, and Frangou (2016). To generate polygenic risk scores for telomere length (PRS-TL), we obtained the genome-wide association study (GWAS) summary statistics from the largest TL GWAS to-date (Codd et al, 2013).…”

Polygenic risk scores of several subtypes of epilepsies in a founder population

“…They can be used to estimate a person's likelihood of displaying any trait with a genetic component. They have been used previously in many common traits and diseases such as heart disease [12][13][14] and more importantly in neurological disorders such as schizophrenia [15][16][17][18] . In the present study, we take advantage of the most recent meta-analysis GWAS metrics 9 to calculate PRS in 522 French-Canadian epilepsy patients divided into seven subtypes.…”