Polygenic Risk Score (PRS Sample Clauses

Polygenic Risk Score (PRS. To enhance risk prediction for onset and outcome of the disorder, GWAS discovery markers can be summarised into a polygenic risk score (PRS). PRS is an individual score consisting of the weighted sum of trait-associated alleles using effect size estimates derived from GWAS summary statistics (Xxxxx & Xxxxxx, 2017). A polygenic risk score is initially calculated on a training sample, before being implemented in a separate independent sample (target sample) (Dudbridge, 2013; Iyegbe, Xxxxxxxx, Xxxxxx, Xxxxxxxx, & Xxxx, 2014). The application of the polygenic risk score technique has had an important impact on genetic research for two main reasons: 1) testing for association between the score and the trait in the replication sample can establish there are indeed associated markers within those contributing to the score; 2) PRS can predict individual risks for diseases, potentially with better discrimination properties than a prediction based on candidate genes. Indeed, since its first applications, PRS analysis has demonstrated significant predictive power in several medical conditions (Xxxxxxxx et al., 2017). The PRS for schizophrenia explained from 3% (Xxxxxxx et al., 2009) to 10% (Xxxxx et al., 2014) of the variance in disease liability, as sample size progressively increased. Although the genetic prediction is expected to improve by increasing the sample size, PRS accuracy has to be considered in light of two important points: 1) genetic factors do not explain the whole liability to the disorder; 2) PRS only measures the contribution of common genetic variants, which do not encompass the totality of genetic risk. A recent study has tested the predictive power of PRS for schizophrenia in a FEP sample (Xxxxxx et al., 2017) finding that it explained around 10% of the variance in liability across cases vs. controls among white Europeans; however, it dropped to ~1% of variance explained among individuals of African ancestry. This is explained by the overrepresentation ( 80%) of individual of European descents in the GWAS used as training samples. (Xxxxxx et al., 2019). Genetic ancestry is estimated by shared genomic signatures (allele frequencies and linkage disequilibrium patterns) due to migration of common ancestors, mutations and recombination, genetic drift, and natural selection, which are different between populations (Xxxxxxxx et al., 2019). Currently, PRS is expected to have some predictive ability for individuals of Asian descents, but to be particula...