Predictive ability of genome-assisted statistical models under various forms of gene action

Mehdi Momen, Ahmad Ayatollahi Mehrgardi, Ayyub Sheikhi, Andreas Kranis, Llibertat Tusell, Gota Morota, Guilherme J.M. Rosa, Daniel Gianola

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Recent work has suggested that the performance of prediction models for complex traits may depend on the architecture of the target traits. Here we compared several prediction models with respect to their ability of predicting phenotypes under various statistical architectures of gene action: (1) purely additive, (2) additive and dominance, (3) additive, dominance, and two-locus epistasis, and (4) purely epistatic settings. Simulation and a real chicken dataset were used. Fourteen prediction models were compared: BayesA, BayesB, BayesC, Bayesian LASSO, Bayesian ridge regression, elastic net, genomic best linear unbiased prediction, a Gaussian process, LASSO, random forests, reproducing kernel Hilbert spaces regression, ridge regression (best linear unbiased prediction), relevance vector machines, and support vector machines. When the trait was under additive gene action, the parametric prediction models outperformed non-parametric ones. Conversely, when the trait was under epistatic gene action, the non-parametric prediction models provided more accurate predictions. Thus, prediction models must be selected according to the most probably underlying architecture of traits. In the chicken dataset examined, most models had similar prediction performance. Our results corroborate the view that there is no universally best prediction models, and that the development of robust prediction models is an important research objective.

Original languageEnglish (US)
Article number12309
JournalScientific reports
Volume8
Issue number1
DOIs
StatePublished - Dec 1 2018

Fingerprint

statistical models
genome
prediction
genes
epistasis
dominance (genetics)
chickens
additive gene effects
genomics
phenotype
loci

ASJC Scopus subject areas

  • General

Cite this

Momen, M., Mehrgardi, A. A., Sheikhi, A., Kranis, A., Tusell, L., Morota, G., ... Gianola, D. (2018). Predictive ability of genome-assisted statistical models under various forms of gene action. Scientific reports, 8(1), [12309]. https://doi.org/10.1038/s41598-018-30089-2

Predictive ability of genome-assisted statistical models under various forms of gene action. / Momen, Mehdi; Mehrgardi, Ahmad Ayatollahi; Sheikhi, Ayyub; Kranis, Andreas; Tusell, Llibertat; Morota, Gota; Rosa, Guilherme J.M.; Gianola, Daniel.

In: Scientific reports, Vol. 8, No. 1, 12309, 01.12.2018.

Research output: Contribution to journalArticle

Momen, M, Mehrgardi, AA, Sheikhi, A, Kranis, A, Tusell, L, Morota, G, Rosa, GJM & Gianola, D 2018, 'Predictive ability of genome-assisted statistical models under various forms of gene action', Scientific reports, vol. 8, no. 1, 12309. https://doi.org/10.1038/s41598-018-30089-2
Momen, Mehdi ; Mehrgardi, Ahmad Ayatollahi ; Sheikhi, Ayyub ; Kranis, Andreas ; Tusell, Llibertat ; Morota, Gota ; Rosa, Guilherme J.M. ; Gianola, Daniel. / Predictive ability of genome-assisted statistical models under various forms of gene action. In: Scientific reports. 2018 ; Vol. 8, No. 1.
@article{2a56a6e10ed5476bad4d7d655224e82d,
title = "Predictive ability of genome-assisted statistical models under various forms of gene action",
abstract = "Recent work has suggested that the performance of prediction models for complex traits may depend on the architecture of the target traits. Here we compared several prediction models with respect to their ability of predicting phenotypes under various statistical architectures of gene action: (1) purely additive, (2) additive and dominance, (3) additive, dominance, and two-locus epistasis, and (4) purely epistatic settings. Simulation and a real chicken dataset were used. Fourteen prediction models were compared: BayesA, BayesB, BayesC, Bayesian LASSO, Bayesian ridge regression, elastic net, genomic best linear unbiased prediction, a Gaussian process, LASSO, random forests, reproducing kernel Hilbert spaces regression, ridge regression (best linear unbiased prediction), relevance vector machines, and support vector machines. When the trait was under additive gene action, the parametric prediction models outperformed non-parametric ones. Conversely, when the trait was under epistatic gene action, the non-parametric prediction models provided more accurate predictions. Thus, prediction models must be selected according to the most probably underlying architecture of traits. In the chicken dataset examined, most models had similar prediction performance. Our results corroborate the view that there is no universally best prediction models, and that the development of robust prediction models is an important research objective.",
author = "Mehdi Momen and Mehrgardi, {Ahmad Ayatollahi} and Ayyub Sheikhi and Andreas Kranis and Llibertat Tusell and Gota Morota and Rosa, {Guilherme J.M.} and Daniel Gianola",
year = "2018",
month = "12",
day = "1",
doi = "10.1038/s41598-018-30089-2",
language = "English (US)",
volume = "8",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Predictive ability of genome-assisted statistical models under various forms of gene action

AU - Momen, Mehdi

AU - Mehrgardi, Ahmad Ayatollahi

AU - Sheikhi, Ayyub

AU - Kranis, Andreas

AU - Tusell, Llibertat

AU - Morota, Gota

AU - Rosa, Guilherme J.M.

AU - Gianola, Daniel

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Recent work has suggested that the performance of prediction models for complex traits may depend on the architecture of the target traits. Here we compared several prediction models with respect to their ability of predicting phenotypes under various statistical architectures of gene action: (1) purely additive, (2) additive and dominance, (3) additive, dominance, and two-locus epistasis, and (4) purely epistatic settings. Simulation and a real chicken dataset were used. Fourteen prediction models were compared: BayesA, BayesB, BayesC, Bayesian LASSO, Bayesian ridge regression, elastic net, genomic best linear unbiased prediction, a Gaussian process, LASSO, random forests, reproducing kernel Hilbert spaces regression, ridge regression (best linear unbiased prediction), relevance vector machines, and support vector machines. When the trait was under additive gene action, the parametric prediction models outperformed non-parametric ones. Conversely, when the trait was under epistatic gene action, the non-parametric prediction models provided more accurate predictions. Thus, prediction models must be selected according to the most probably underlying architecture of traits. In the chicken dataset examined, most models had similar prediction performance. Our results corroborate the view that there is no universally best prediction models, and that the development of robust prediction models is an important research objective.

AB - Recent work has suggested that the performance of prediction models for complex traits may depend on the architecture of the target traits. Here we compared several prediction models with respect to their ability of predicting phenotypes under various statistical architectures of gene action: (1) purely additive, (2) additive and dominance, (3) additive, dominance, and two-locus epistasis, and (4) purely epistatic settings. Simulation and a real chicken dataset were used. Fourteen prediction models were compared: BayesA, BayesB, BayesC, Bayesian LASSO, Bayesian ridge regression, elastic net, genomic best linear unbiased prediction, a Gaussian process, LASSO, random forests, reproducing kernel Hilbert spaces regression, ridge regression (best linear unbiased prediction), relevance vector machines, and support vector machines. When the trait was under additive gene action, the parametric prediction models outperformed non-parametric ones. Conversely, when the trait was under epistatic gene action, the non-parametric prediction models provided more accurate predictions. Thus, prediction models must be selected according to the most probably underlying architecture of traits. In the chicken dataset examined, most models had similar prediction performance. Our results corroborate the view that there is no universally best prediction models, and that the development of robust prediction models is an important research objective.

UR - http://www.scopus.com/inward/record.url?scp=85051703874&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051703874&partnerID=8YFLogxK

U2 - 10.1038/s41598-018-30089-2

DO - 10.1038/s41598-018-30089-2

M3 - Article

C2 - 30120288

AN - SCOPUS:85051703874

VL - 8

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 12309

ER -