Lowers barriers to entry
Shared infrastructure makes it easier for researchers to access genetically informative data without navigating separate access processes, data agreements, and analysis environments for every contributing cohort.
About the ETAF Data Commons
The scientific case for shared infrastructure — why family-based cohorts need a common resource, what the ETAF Data Commons aims to build, and what becomes scientifically possible at scale.
Why This Is Needed
The scientific case for shared infrastructure
Twin, adoption, and extended-family cohorts have delivered decades of insight into genetic and environmental influences on human behavior, health, development, and social outcomes. Yet these datasets remain scattered across institutions, governed independently, difficult for outside researchers to access, and often measured using different instruments.
Many of the most important questions in behavioral and biomedical science require scale, family structure, longitudinal measurement, and genomic data together — conditions that individual cohorts and conventional population biobanks typically cannot meet alone. Shared infrastructure can substantially increase the scientific return on decades of cohort investment.
Harmonizes phenotypes
Phenotype harmonization will be pursued where scientifically appropriate, and cohort-specific measures will be preserved when harmonization would obscure meaningful differences or reduce scientific value.
Increases power and supports replication
Combining independently collected cohorts increases statistical power and enables systematic replication — two conditions essential for producing findings that are credible, robust, and cumulative.
Supports new family-based genomic methods
Modern methods — within-family GWAS, indirect genetic effects models, extended pedigree analyses — require large samples of families with genomic data that no single cohort can currently provide.
What We're Building
A secure, integrated research resource
The ETAF Data Commons is intended to provide shared computational and governance infrastructure for integrating family-based cohorts in a secure, controlled-access environment. All elements described below are in development; nothing has been finalized or deployed.
Contributing Cohorts
Twin, adoption, extended family, and related genetically informative studies worldwide
Intake & Metadata
Study documentation, family-structure metadata, and consent constraint mapping
Phenotype Harmonization
Cross-cohort harmonization where appropriate; cohort-specific measures preserved
Genomic Processing
QC, imputation, and integration of genomic data where available; DNA pathways explored
Secure Cloud Platform
Controlled-access analysis environment; no broad individual-level data download
PlannedApproved Researchers & Discovery
Qualified investigators with approved projects conduct analyses and generate new knowledge
All stages are under development. Future access will be project-based, governed by data-use agreements, and subject to cohort-specific consent constraints. Individual-level data will not be broadly downloadable.
Scientific Opportunities
What becomes possible at scale
Combining family-based cohorts with harmonized phenotypes and genomic data enables scientific approaches that are not feasible in isolated datasets. The following represent a sample of anticipated use cases.
Within-family genomics
Sibling and twin-pair analyses to estimate causal genetic effects, control for passive gene-environment correlation, and test causal genetic hypotheses.
Adoption & rearing environment
Adoption designs offer strong leverage on causal effects of rearing environments, independent of genetic transmission from biological parents.
Twin & sibling comparisons
Classic and extended twin designs to decompose genetic and environmental variance, test GxE interactions, and evaluate biometric model assumptions.
Extended pedigree & intergenerational
Multi-generational data to study intergenerational transmission, indirect genetic effects, and the developmental origins of health and behavior.
Gene-environment correlation & interplay
Family-based designs offer powerful approaches to distinguish active, reactive, and passive gene-environment correlations and identify GxE interactions.
Assortative mating
Spousal data and family pedigrees to study mate selection, phenotypic and genetic resemblance between partners, and implications for population genetic structure.
Developmental & life-course research
Longitudinal phenotyping across the life span to study how genetic and environmental influences on traits change from childhood through adulthood.
Replication & method benchmarking
Cross-cohort replication of key findings, evaluation of statistical methods under different design assumptions, and construction of multi-cohort reference datasets.