A data harmonization pipeline to leverage external controls and boost power in GWAS.

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• External controls enhance genome-wide association studies (GWAS) by increasing sample size and diversity.
• Aggregating external controls is challenging due to batch effects, genotyping errors, and platform differences.
• These challenges can lead to spurious results if not properly addressed.

Purpose of the Study:

• To develop and validate a unified data harmonization pipeline for aggregating external control genotype data.
• To improve the power of GWAS by leveraging large, diverse, harmonized control cohorts.
• To demonstrate the pipeline's effectiveness in a Crohn's disease GWAS.

Main Methods:

• Proposed an iterative quality control and imputation pipeline implemented before and after data merging.
• Applied the pipeline to aggregate 27,517 European control samples from 16 collections in dbGaP.
• Conducted a GWAS of Crohn's disease using the harmonized external controls.

Main Results:

• The harmonization pipeline successfully aggregated genotype data from multiple sources, yielding 27,517 European controls.
• GWAS using harmonized controls showed increased power compared to using cohort samples alone.
• The procedure generated summary statistics free of significant batch effects.

Conclusions:

• The proposed harmonization pipeline effectively addresses challenges in aggregating external control data for GWAS.
• This approach enhances the power of genetic association studies and reduces spurious findings.
• The pipeline is applicable to other research requiring aggregated individual-level genotype data.