Generalized Odds Ratio in Meta-Analysis of Genetic Studies
Author Information
Author(s): Tiago V. Pereira, Regina C. Mingroni-Netto
Primary Institution: Centro de Estudos do Genoma Humano, Universidade de São Paulo
Hypothesis
The generalized odds ratio (GOR) can serve as a model-free measure for association studies involving bi- and tri-allelic polymorphisms.
Conclusion
The GOR may be slightly more powerful for tri-allelic variants, especially when susceptibility alleles are less common.
Supporting Evidence
- The GOR performs similarly to standard models for bi-allelic polymorphisms.
- The GOR may enhance power for tri-allelic variants with low-frequency alleles.
- Bias in effect size estimates increases with the minor allele frequency.
Takeaway
This study looks at a new way to analyze genetic data that might help find important genetic links to diseases like Alzheimer's.
Methodology
Monte Carlo simulations were used to investigate type-I error rates, power, and bias in meta-analyses using the GOR.
Potential Biases
The GOR may overestimate the true underlying increase in effect size for common variants.
Limitations
The GOR may provide inflated effects for bi-allelic variants following a multiplicative model of action.
Participant Demographics
The study involved 1411 participants, including 961 cases and 560 controls.
Statistical Information
P-Value
2.14 × 10-32
Confidence Interval
95% CI for odds ratios ranged from 0.31 to 5.57.
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website