Related Experiment Video
Updated: Jan 2, 2026
08:03
Heuristic Mining of Hierarchical Genotypes and Accessory Genome Loci in Bacterial Populations
Published on: December 7, 2021
2.7K
A hidden Markov model for haplotype inference for present-absent data of clustered genes using identified haplotypes
Jihua Wu1, Guo-Bo Chen2, Degui Zhi1
1Section on Statistical Genetics, Department of Biostatistics, University of Alabama at Birmingham Birmingham, AL, USA.
Frontiers in Genetics
|August 28, 2014
Summary
A new hidden Markov model (HMM) method improves killer cell immunoglobulin-like receptor (KIR) gene haplotype inference by incorporating known patterns. This approach enhances accuracy, especially when dealing with ambiguous gene copy number data.
Area of Science:
- Genetics
- Immunogenetics
- Computational Biology
Background:
- Killer cell immunoglobulin-like receptor (KIR) gene genotyping often results in present/absent data, leading to ambiguity in copy number determination.
- This ambiguity complicates accurate haplotype inference for KIR genes, which are crucial for immune system function.
- Tight linkage disequilibrium among clustered KIR genes allows for the use of known haplotypes and patterns to aid inference.
Purpose of the Study:
- To develop a novel hidden Markov model (HMM) based method for KIR gene haplotype inference.
- To incorporate previously identified haplotypes and partial haplotype patterns into the inference process.
- To evaluate the performance of the new method against existing approaches.
Main Methods:
- Development of a hidden Markov model (HMM) that integrates known haplotype and partial haplotype patterns.
- Comparison of the HMM method with an expectation maximization (EM) based method using extensive simulations for KIR genes.
- Benchmarking against methods that do not utilize prior haplotype information, including HPALORE (EM) and MaCH (HMM).
Main Results:
- The new HMM-based method demonstrated superior performance in haplotype assignment and frequency estimation compared to a previous EM-based method, particularly under specific simulation conditions.
- The incorporation of identified haplotypes and partial patterns significantly improved the accuracy of haplotype inference.
- The developed software package, HaploHMM, showed improved accuracy over methods not leveraging prior haplotype information.
Conclusions:
- The novel HMM-based method effectively addresses the challenges in KIR gene haplotype inference caused by ambiguous copy number data.
- Utilizing known haplotypes and partial patterns is a valuable strategy for enhancing the accuracy of genetic haplotype inference.
- The HaploHMM software provides an improved tool for complex genetic analyses involving clustered genes like KIR.
