Tags

Type your tag names separated by a space and hit enter

Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for estimating genetic parameters and breeding values.
Genet Sel Evol. 2019 Dec 10; 51(1):73.GS

Abstract

BACKGROUND

Hamiltonian Monte Carlo is one of the algorithms of the Markov chain Monte Carlo method that uses Hamiltonian dynamics to propose samples that follow a target distribution. The method can avoid the random walk behavior to achieve a more effective and consistent exploration of the probability space and sensitivity to correlated parameters, which are shortcomings that plague many Markov chain Monte Carlo methods. However, the performance of Hamiltonian Monte Carlo is highly sensitive to two hyperparameters. The No-U-Turn Sampler, an extension of Hamiltonian Monte Carlo, was recently introduced to automate the tuning of these hyperparameters. Thus, this study compared the performances of Gibbs sampling, Hamiltonian Monte Carlo, and the No-U-Turn Sampler for estimating genetic parameters and breeding values as well as sampling qualities in both simulated and real pig data. For all datasets, we used a pedigree-based univariate linear mixed model.

RESULTS

For all datasets, the No-U-Turn Sampler and Gibbs sampling performed comparably regarding the estimation of heritabilities and accuracies of breeding values. Compared with Gibbs sampling, the estimates of effective sample sizes for simulated and pig data with the No-U-Turn Sampler were 3.2 to 22.6 and 3.5 to 5.9 times larger, respectively. Autocorrelations decreased more quickly with the No-U-Turn Sampler than with Gibbs sampling. When true heritability was low in the simulated data, the skewness of the marginal posterior distributions with the No-U-Turn Sampler was smaller than that with Gibbs sampling. The performance of Hamiltonian Monte Carlo for sampling quality was inferior to that of No-U-Turn Sampler in the simulated data. Moreover, Hamiltonian Monte Carlo could not estimate genetic parameters because of difficulties with the hyperparameter settings with pig data.

CONCLUSIONS

The No-U-Turn Sampler is a promising sampling method for animal breeding because of its good sampling qualities: large effective sample sizes, low autocorrelations, and low skewness of marginal posterior distributions, particularly when heritability is low. Meanwhile, Hamiltonian Monte Carlo failed to converge with a simple univariate model for pig data. Thus, it might be difficult to use Hamiltonian Monte Carlo for usual complex models in animal breeding.

Authors+Show Affiliations

Institute of Livestock and Grassland Science, NARO, 2 Ikenodai Tsukuba, Ibaraki, 3050901, Japan. mtnishio@affrc.go.jp.Institute of Livestock and Grassland Science, NARO, 2 Ikenodai Tsukuba, Ibaraki, 3050901, Japan.

Pub Type(s)

Evaluation Study
Journal Article

Language

eng

PubMed ID

31823719

Citation

Nishio, Motohide, and Aisaku Arakawa. "Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for Estimating Genetic Parameters and Breeding Values." Genetics, Selection, Evolution : GSE, vol. 51, no. 1, 2019, p. 73.
Nishio M, Arakawa A. Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for estimating genetic parameters and breeding values. Genet Sel Evol. 2019;51(1):73.
Nishio, M., & Arakawa, A. (2019). Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for estimating genetic parameters and breeding values. Genetics, Selection, Evolution : GSE, 51(1), 73. https://doi.org/10.1186/s12711-019-0515-1
Nishio M, Arakawa A. Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for Estimating Genetic Parameters and Breeding Values. Genet Sel Evol. 2019 Dec 10;51(1):73. PubMed PMID: 31823719.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Performance of Hamiltonian Monte Carlo and No-U-Turn Sampler for estimating genetic parameters and breeding values. AU - Nishio,Motohide, AU - Arakawa,Aisaku, Y1 - 2019/12/10/ PY - 2019/06/06/received PY - 2019/11/27/accepted PY - 2019/12/12/entrez PY - 2019/12/12/pubmed PY - 2020/4/17/medline SP - 73 EP - 73 JF - Genetics, selection, evolution : GSE JO - Genet Sel Evol VL - 51 IS - 1 N2 - BACKGROUND: Hamiltonian Monte Carlo is one of the algorithms of the Markov chain Monte Carlo method that uses Hamiltonian dynamics to propose samples that follow a target distribution. The method can avoid the random walk behavior to achieve a more effective and consistent exploration of the probability space and sensitivity to correlated parameters, which are shortcomings that plague many Markov chain Monte Carlo methods. However, the performance of Hamiltonian Monte Carlo is highly sensitive to two hyperparameters. The No-U-Turn Sampler, an extension of Hamiltonian Monte Carlo, was recently introduced to automate the tuning of these hyperparameters. Thus, this study compared the performances of Gibbs sampling, Hamiltonian Monte Carlo, and the No-U-Turn Sampler for estimating genetic parameters and breeding values as well as sampling qualities in both simulated and real pig data. For all datasets, we used a pedigree-based univariate linear mixed model. RESULTS: For all datasets, the No-U-Turn Sampler and Gibbs sampling performed comparably regarding the estimation of heritabilities and accuracies of breeding values. Compared with Gibbs sampling, the estimates of effective sample sizes for simulated and pig data with the No-U-Turn Sampler were 3.2 to 22.6 and 3.5 to 5.9 times larger, respectively. Autocorrelations decreased more quickly with the No-U-Turn Sampler than with Gibbs sampling. When true heritability was low in the simulated data, the skewness of the marginal posterior distributions with the No-U-Turn Sampler was smaller than that with Gibbs sampling. The performance of Hamiltonian Monte Carlo for sampling quality was inferior to that of No-U-Turn Sampler in the simulated data. Moreover, Hamiltonian Monte Carlo could not estimate genetic parameters because of difficulties with the hyperparameter settings with pig data. CONCLUSIONS: The No-U-Turn Sampler is a promising sampling method for animal breeding because of its good sampling qualities: large effective sample sizes, low autocorrelations, and low skewness of marginal posterior distributions, particularly when heritability is low. Meanwhile, Hamiltonian Monte Carlo failed to converge with a simple univariate model for pig data. Thus, it might be difficult to use Hamiltonian Monte Carlo for usual complex models in animal breeding. SN - 1297-9686 UR - https://www.unboundmedicine.com/medline/citation/31823719/Performance_of_Hamiltonian_Monte_Carlo_and_No_U_Turn_Sampler_for_estimating_genetic_parameters_and_breeding_values_ L2 - https://gsejournal.biomedcentral.com/articles/10.1186/s12711-019-0515-1 DB - PRIME DP - Unbound Medicine ER -