Sjoerd van Alten

Thrilled to see this joint work out!
Big thanks to my amazing coauthors: Silvia Barcellos, Leandro Carvalho, Titus Galama, and Marina Aguiar Palma. (8/8)

Key takeaway:
Even variation rooted in nature—our genes—exerts much of its influence through nurture. (7/8)

We quantify these three channels and find:
- Direct genetic transmission and genetic nurture both play substantial roles
- Assortative mating is comparatively minor
- For wealth outcomes, genetic nurture > direct transmission (6/8)

This shows parental genes matter not only through direct inheritance but also via:
- Genetic nurture – how parental genes shape the child’s environment
- Assortative mating – non-random partnering patterns (5/8)

Our findings: "next-generation" effects of parental PGI on children's outcomes are surprisingly large, as compared to "same-generation" effects (the effects of the parent's PGI on their own socioeconomic status). (4/8)

To isolate causality, we exploit the natural randomization of genes at conception, conditioning on grandparents’ PGIs.
This lets us separate pure genetic transmission from environmental effects. (3/8)

Using a unique linkage of genetic data from Lifelines_NL and administrative records from Centraal Bureau voor de Statistiek (CBS), we ask:
How do a parent’s genes associated with educational attainment—measured by a polygenic index (PGI)—affect their children’s socioeconomic outcomes? (2/8)

A Chip Off the Old Block? Genetics and the Intergenerational Transmission of Socioeconomic Status Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

Proud to share our new @nber working paper on how genetics shape the intergenerational transmission of socioeconomic status in the Netherlands. 🧵(1/8)

www.nber.org/papers/w34208

Genetics play a role on the persistence of socioeconomic across generations: one generation's genetics significantly impacts the education, income, and wealth of the next, from Sjoerd van Alten, Silvia H. Barcellos, Leandro Carvalho, Titus J. Galama, and Marina ... https://www.nber.org/papers/w34208

Call for abstracts: genetics, economic & social issues.

We're hosting a 1-day workshop on using genetic data to examine economic & social issues on 12th December at UCL’s Social Research Institute. More info & submission at link below #genetics #socialscience #economics #cohort

bit.ly/41EnPmu

📌

📌

📌

Extremely excited to share the first effort of the Revived Genomics of Personality Consortium: A highly-powered, comprehensive GWAS of the Big Five personality traits in 1.14 million participants from 46 cohorts. www.biorxiv.org/content/10.1...

📌

📌

📌

📌

📣 I’m delighted to share a new working paper that’s been years in the making:

🧬 “ #Gene × #Environment Interactions: Polygenic Scores and the Impact of an Early Childhood Intervention in Colombia”

👉🏻 Available here as @hceconomics.bsky.social WP: humcap.uchicago.edu/RePEc/hka/wp...

📌

📌

I am recruiting a quantitative/computational postdoc to my group at UCLA. This is a great opportunity to work on foundational theory, methods, and software in statistical genetics. Link to apply: recruit.apo.ucla.edu/JPF10275. Please repost!

Agreed! The opportunity for follow-up analyses is endless. One thing I forgot to mention here is that these weights are available in the Returns Catalogue to any researchers who use the UKB, under application# 55154: biobank.ndph.ox.ac.uk/ukb/app.cgi?...

Many thanks to my amazing co-authors: Ben Domingue, Jessica Faul, Titus Galama, and Andries Marees. This paper has been a 4-year long journey and I am so happy to finally see it out!

Overall, the message is clear: volunteer bias matters to GWAS results and downstream analyses. The extent to which it matters is phenotype-specific. The community should work on creating population-representative weights for various cohorts and incorporate these in GWAS.

WGWAS may also result in different bio annotations (as estimated in MAGMA). For example, the GWAS results for breast cancer show no enriched pathways. The WGWAS results are expressed in the fallopian tube, uterus, ovary, and Artery Tibial (Figure 3).

Furthermore, we find evidence that weighting GWAS results pushes the intercept of LD-score regression closer to 1, which indicates that weighting might also shield against bias due to population stratification.

WGWAS also resulted in larger SNP-based heritabilities for 7 out of the 10 phenotypes (Table 3). For example, Years of education shows a SNP-based heritability of 14.8% in GWAS, and 17.8% in WGWAS.

Note also that the effective sample size shrinks from 376,900 in GWAS to 143,222 in WGWAS averaged over all phenotypes, a shrinkage of 62%. Hence, representative samples would increase the power of GWAS, as power reduces when taking volunteer bias into account.

Next, we compare GWAS and WGWAS associations genomewide. The genetic correlation between GWAS and WGWAS results is lower than one for 6 out of 10 phenotypes. The lowest congruence between GWAS and WGWAS is found for Type 1 Diabetes (rG=0.66) and Breast Cancer (rG = 0.80). (Table 2)