Comparing Methods for Analyzing Genetic Data Over Time
Author Information
Author(s): Yang Qiong, Chazaro Irmarie, Cui Jing, Guo Chao-Yu, Demissie Serkalem, Larson Martin, Atwood Larry D, Cupples L Adrienne, DeStefano Anita L
Primary Institution: Boston University
Hypothesis
Can different statistical methods effectively analyze longitudinal cholesterol data to identify genetic influences?
Conclusion
Univariate methods can identify genes affecting cholesterol levels, but multivariate methods provide better insights into how these effects change with age.
Supporting Evidence
- Univariate methods detected several genes affecting cholesterol levels.
- The multivariate approach provided additional insights into how gene effects change with age.
- Different methods yielded varying heritability estimates for cholesterol levels.
Takeaway
This study looked at different ways to analyze cholesterol data over time to see how genes affect it. Some methods are easier but miss important details, while others are more complex but give better information.
Methodology
The study compared two univariate methods and one multivariate method to analyze cholesterol levels using variance components models.
Limitations
The multivariate approach involves complex modeling and heavy computation, which may limit its practical application.
Participant Demographics
Subjects were aged between 20 and 93, with varying numbers of cholesterol measurements.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website