Clustering Algorithm for Haplotype Studies
Author Information
Author(s): Lee Mei-Hsien, Tzeng Jung-Ying, Huang Su-Yun, Hsiao Chuhsing Kate
Primary Institution: Taipei Municipal University of Education
Hypothesis
Can an evolution-guided clustering algorithm improve haplotype-based association studies in families?
Conclusion
The proposed clustering procedure improves the power of association tests by reducing haplotype dimensionality and accounting for phase ambiguity.
Supporting Evidence
- The proposed procedure incorporates evolutionary information and reduces the number of haplotypes for analysis.
- Simulation studies showed that the new method is more powerful than existing family-based association tests.
- The clustering approach helps to manage haplotype phase ambiguity and transmission uncertainty.
Takeaway
This study created a new way to group genetic data to make it easier to find links between genes and diseases, especially when dealing with complex family data.
Methodology
The study used family genotype data and combined a clustering scheme with a likelihood ratio statistic to test associations between phenotypes and haplotype variants.
Limitations
The choice of core haplotypes based on cumulative frequency is somewhat arbitrary and may not effectively reduce dimensionality in large sample sizes.
Participant Demographics
The study involved families with children, with a focus on genetic data from these families.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website