Genome-wide Association Mapping Using Mixed Models
Author Information
Author(s): Keyan Zhao, Magnus Nordborg, Paul Marjoram
Primary Institution: University of Southern California
Hypothesis
Can mixed-model methods effectively reduce false-positive rates in genome-wide association studies?
Conclusion
The study found that while mixed models can be used to analyze genome-wide SNP data, they did not significantly reduce the false-positive rate in the simulated data.
Supporting Evidence
- Mixed models can effectively analyze genome-wide SNP data.
- The method did not significantly reduce the false-positive rate in the simulated data.
- The study found strong signals in the correct locations for the analyzed data.
Takeaway
This study looked at how to find real genetic signals in data without getting tricked by false alarms. They found that their method works, but it didn't help as much as they hoped.
Methodology
The study used a generalized linear mixed-model to analyze SNP data from affected sib pairs and unrelated controls.
Potential Biases
The method may exaggerate under-dispersion of p-values, which has not been documented previously.
Limitations
The simulation scheme used resulted in data with limited relatedness, which may not provide an ideal test for the method.
Participant Demographics
The study analyzed data from 1500 affected sib pairs and 2000 unrelated control subjects.
Statistical Information
P-Value
1 × 10^-4
Statistical Significance
p<0.0001
Want to read the original?
Access the complete publication on the publisher's website