Аннотация:In the recent decade polygenic risk score (PRS) approach became of high relevance in order to predict individual risk at developing multifactorial traits and diseases. However, PRS studies to predict spatial abilities, which determine individual life and professional success, remain scarce. Therefore, we aimed to construct PRS models including SNPs and socio-demographic factors as predictors of individual trajectory in spatial abilities in mentally healthy young adults.The study included 1011 mentally healthy individuals aged 18-25 years of Caucasian origin (535 Russians, 231 Tatars, 160 Udmurts, and 85 of mixed ethnicity) from Russia. Spatial ability was assessed using “shape rotation” technique implemented in the cognitive tests battery (InLab, Goldsmiths, University of London). We selected and genotyped 20 SNPs in the NGF, NRXN1, KIBRA, NRG1, BDNF, GRIN2B, APOE, SNAP25, IL1b, IL1A, CRP, TNF, P2X7R, C3AR1, DNMT1, DNMT3a, DNMT3b genes, which reported GWAS-based association with cognitive traits. Genotypes and 20 social parameters served as predictors in MLR models, which determined regression coefficients in the training sample (N=711) served the basis for PRS calculation in the testing sample (N=300) (PLINK v.1.09, R).The most significant weighted PRS model demonstrated higher sensitivity to predict spatial ability in the testing sample while adjusting for sex and explained up to 9.13% of variance (β=2.11, p=1.52e-07). According to identified model, men demonstrated higher ability to 3D mental rotation than women (β = 3.71, Р = 8.32e-05). Constructed model evidence in significant effects of examined genetic variants and sex in predicting individual differences in spatial abilities.