A method for improving cross-population polygenic risk scores is reported in Nature Genetics this week, offering a new tool for addressing health disparities between European and non-European populations. Despite the potential of polygenic risk scores in identifying disease risk, most use European training data, resulting in lower accuracy when they are applied to individuals of non-European ancestry. To address this, a team led by scientists from Harvard University developed PolyPred, a combination of a published predictor and a novel predictor that leverages functionally informed fine-mapping to estimate causal effects rather than tagging effects to address linkage disequilibrium differences. When applied to 49 diseases and complex traits in four populations within the UK Biobank, Biobank Japan, and Uganda-APCDR, PolyPred improved polygenic prediction accuracy in non-European populations compared with other state-of-the-art prediction methods.
A large-scale exome sequencing project in bipolar disorder (BD) patients published in Nature Genetics this week has uncovered a new risk gene shared with schizophrenia, suggesting that researchers are beginning to discover true risk factors underlying psychiatric disease. In the study, members of the international Bipolar Exome initiative analyzed whole-exome sequencing data for 13,933 patients with BD and 14,422 controls. They find an excess of ultra-rare protein-truncating variants in BD patients among genes under strong evolutionary constraint in both major subtypes of the disease, a result firmly established in schizophrenia and other early-onset neurodevelopmental disorders. The scientists also observe that shared risk for BD and schizophrenia is present in both common and damaging ultra-rare variation, with one gene in particular — AKAP11 — emerging as a definitive risk gene in both conditions. "Overall, the current evidence suggests gene discovery in BD is on a similar trajectory to schizophrenia, where increased sample sizes and further collaborative efforts will inevitably lead to biologically meaningful risk genes and pathways underlying BD risk," the researchers conclude.
A number of previously unidentified gene variants associated with facial shape variation in European and Asian individuals are reported in a new study appearing in this week's Nature Genetics. To date, studies investigating the genetic basis of facial morphology have been largely limited to European populations. In this week's report, a team led by Fudan University scientists performed a genome-wide association study based on a large collection of three-dimensional facial images from the Han Chinese population, uncovering 244 variants in 166 loci — 62 of which are new — associated with typical-range facial variation. The researchers further identified 13 variants relate to facial shape differences between European and East Asian populations, with evolutionary analyses suggesting that the difference in nose shape between the two populations is caused by a directional selection, due mainly to a local adaptation in Europeans.