Incorporating polygenic liability and family history for predicting psychiatric diseases in the Taiwan biobank.

Abstract

Background: This study investigated the interplay between molecular measures of polygenic risk score (PRS) and conventional measures of family history (FH) on the risk of four psychiatric disorders: schizophrenia (SCZ), bipolar disorder (BPD), major depressive disorder (MDD), and obsessive-compulsive disorder (OCD) in community samples of East Asian populations. We examined the individual and joint associations and relative contributions of PRS and FH and evaluated the potential of combining transdiagnostic PRSs and FHs to improve risk prediction.

Methods: The genotyping of 106,581 unrelated participants from the Taiwan Biobank was linked to the National Health Insurance Research Database to retrieve information on ICD-defined diseases and FH. A logistic regression model was used to examine the association between PRS and FH in fathers, mothers, and siblings with a risk of psychiatric disorders.

Results: The PRS for SCZ, BPD, MDD, and OCD explained 2.0%, 0.4%, 0.6%, and 0.6%, respectively, and FH explained 1.3%, 1.4%, 2.3%, and 3.4%, respectively, of the variance in the corresponding disease. Incorporating PRS and FH increased the explained variances in SCZ, BPD, MDD, and OCD by 3.2%, 1.7%, 2.8%, and 4.1%, respectively. The effect sizes for PRS and FH in the PRS/FH alone and PRS-FH combined models were generally similar. Simultaneously incorporating the four PRSs and FHs increased the explained variances of SCZ, BPD, MDD, and OCD to 4.7%, 4.7%, 3.3%, and 7.3%, respectively.

Conclusions: PRS and FH provide independent and complementary information for the identification of psychiatric disorders. The incorporation of transdiagnostic PRSs and FHs improved risk identification.