Table 2. Associations between child and parent polygenic scores with phenotypes (education, n = 1095 trios; CRP, n = 942 trios).

SEs in parentheses. Independent associations represent regression models with only a single parent polygenic score (PGS) variable included; adjusted associations represent regression models with all three PGS variables. All models control for the first 20 principal components of ancestry. P values obtained from tests of seemingly unrelated regression on the child PGS coefficients between independent and coadjusted models. SEs for attenuation were obtained through bootstrapping with 1000 replications.

	Independent associations	Adjusted associations	P value for difference	Attenuation in child PGS coefficient
Education
Education PGS all SNPs
Child PGS	0.340 (0.028)	0.223 (0.041)	1.9 × 10-4	34.4% (9.3)
Mother PGS	0.261 (0.029)	0.110 (0.036)
Father PGS	0.232 (0.030)	0.099 (0.034)
Education PGS GWAS SNPs
Child PGS	0.129 (0.029)	0.051 (0.042)	0.009	60.5% (35.0)
Mother PGS	0.072 (0.030)	0.035 (0.036)
Father PGS	0.142 (0.029)	0.111 (0.036)
CRP
CRP GWAS SNPs
Child PGS	0.219 (0.033)	0.192 (0.043)	0.330	12.4% (13.3)
Mother PGS	0.079 (0.033)	−0.002 (0.037)
Father PGS	0.144 (0.032)	0.056 (0.037)