Skip to main content

1000 Genomes

Purpose​

Run Fed-GWAS on a real genotype workflow using a prepared 1000 Genomes chr22 subset and generated phenotypes.

Data​

Follow the experiment-specific instructions in experiments/real_world/1000genomes/README.md. The raw and generated PLINK data are not committed.

Command​

Use the real-world config path:

flwr run . local-simulation --stream --run-config \
  'config_path="experiments/real_world/1000genomes/configs"'

For deployment or cluster runs, point each SuperNode at the corresponding center_x/config.yaml under the same config directory.

Expected output​

  • Federated QC, KING, and LR outputs under the configured real-world results directory.
  • Optional centralized baseline and evaluation reports from experiments/tools/evaluation/.
  • Manuscript figures and tables under experiments/real_world/1000genomes/manuscript/ when generated.