Polygenic score predictive power for educational attainment increases with discovery sample size and is projected to approach the SNP heritability bound¶

Claim · All claims
Evidence · 1 study (3 entries) · review r2 · q2
Currency · newest study 2017 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? PGSs for educational attainment built from discovery samples of about 400,000 explain about R2 ≈ 7% of variance in independent samples. → David Cesarini and Peter M. Visscher 2017 q2 i? An analytical projection indicates a PGS built from a 2M-individual discovery sample would come very close to the SNP heritability upper bound. → David Cesarini and Peter M. Visscher 2017 (2) q2 i? Published estimates suggest common variants account for around 25% of variation in educational attainment, the theoretical ceiling for PGSs built from common variants. → David Cesarini and Peter M. Visscher 2017 (3)

Evidence¶

David Cesarini and Peter M. Visscher 2017¶

David Cesarini and Peter M. Visscher. (2017). Genetics and educational attainment. npj Science of Learning 2:4. https://doi.org/10.1038/s41539-017-0005-6

q2 · i? · review · r2

Analysis of genotyped Health and Retirement Study respondents of European ancestry (Fig. 2): as discovery samples grew to N ≈ 400,000 for EduYears, PGS predictive power rose to "about R2 ≈ 7%".

"As sample sizes have increased to N ≈ 300,000 (BMI) and N ≈ 400,000 (EduYears), the predictive power of PGSs has increased to about R2 ≈ 7%."

David Cesarini and Peter M. Visscher 2017 (2)¶

David Cesarini and Peter M. Visscher. (2017). Genetics and educational attainment. npj Science of Learning 2:4. https://doi.org/10.1038/s41539-017-0005-6

q2 · i? · review · r2

Analytical projection from Eq. 3 using M = 70,000 and R2 and N from the largest published GWAS: a PGS from a 2M discovery sample "comes very close to the theoretical upper bound given by the SNP heritability".

"The projection suggests the predictive power of a polygenic score constructed using weights from a discovery sample of 2M individuals comes very close to the theoretical upper bound given by the SNP heritability."

David Cesarini and Peter M. Visscher 2017 (3)¶

David Cesarini and Peter M. Visscher. (2017). Genetics and educational attainment. npj Science of Learning 2:4. https://doi.org/10.1038/s41539-017-0005-6

q2 · i? · review · r2

Review of published SNP-heritability estimates: common variants account for "around 25% of variation in traits such as BMI and educational attainment", the bound for PGSs built from common variants.

"Published estimates suggest common variants account for about 50% of variation in human height, 21 and around 25% of variation in traits such as BMI and educational attainment."

Discussion¶

Related Claims¶

• EduYears GPS predicts more variance in GCSE scores (about 10%) than in its GWA target trait, adult years of education (about 4%) — related
• Polygenic score predictions of education strengthen as GWAS discovery sample sizes increase — possibly the same claim (merge candidate)
• A polygenic score for educational attainment explains about 15% of individual differences in out-of-sample prediction among people of European descent, with a correlation of approximately 0.4 with outcomes — a narrower finding that bears on this claim
• Educational achievement is highly heritable across school years and subjects, with twin heritability around 60% and SNP heritability around 30% — related
• Polygenic score predictions are reduced in African American relative to European ancestry samples, significantly so only for educational attainment — related
• Sample ancestry significantly moderates polygenic score predictions of educational attainment, with stronger predictions in European-ancestry samples — related
• EduYears polygenic scores predict achievement increasingly with age (about 4% at KS1 to 10% at GCSE) and show largely stable effects, with less than 1% age-specific prediction — related
• A polygenic score for educational attainment positively predicts reading and speaking skills in early childhood, later scholastic achievement, and adult economic security — related