Phenotype harmonization and cross-study collaboration in GWAS consortia: the GENEVA experience

Area of Science:

• Genetics
• Bioinformatics
• Epidemiology

Background:

• Genome-wide association study (GWAS) consortia increase sample size but face challenges with phenotype heterogeneity across studies.
• Combining previously collected phenotype data from different instruments and timeframes is a complex process.
• Phenotype harmonization is essential for successful genetic association studies and gene-environment interaction investigations.

Purpose of the Study:

• To describe strategies and pitfalls for combining phenotype data from varying studies.
• To provide guidance for GWAS consortia on phenotype harmonization using the Gene Environment Association Studies (GENEVA) consortium as an example.
• To highlight key issues in data sharing across disparate studies for genetic research.

Main Methods:

• Identifying common phenotypes across multiple studies.
• Assessing the feasibility of cross-study analysis for identified phenotypes.
• Developing common data definitions and applying appropriate harmonization algorithms.
• Considering factors like genotyping timeframes and imputation of genotype data.

Main Results:

• Phenotype harmonization involves identifying commonalities, assessing feasibility, defining standards, and applying algorithms.
• Challenges include diverse disease endpoints and varying data collection instruments across studies.
• The GENEVA consortium's experience illustrates practical approaches and potential pitfalls in data harmonization.
• Data sharing and collaboration policies are vital for successful multi-site genetic studies.

Conclusions:

• Effective phenotype harmonization is critical for the success of large-scale GWAS and gene-environment interaction studies.
• The strategies and lessons learned from the GENEVA consortium can guide other collaborative efforts.
• Promoting data sharing and collaboration is essential for advancing genetic research across diverse studies.