Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics

Abstract Background Likelihood-based inferences such as variance components estimation and hypothesis testing need logarithms of the determinant (log-determinant) of high dimensional matrices. Calculating the log-determinant is memory and time-consuming, making it impossible to perform likelihood-ba...

Full description

Saved in:
Bibliographic Details
Main Authors: Matias Bermann, Alejandra Alvarez-Munera, Andres Legarra, Ignacio Aguilar, Ignacy Misztal, Daniela Lourenco
Format: Article
Language:deu
Published: BMC 2025-08-01
Series:Genetics Selection Evolution
Online Access:https://doi.org/10.1186/s12711-025-00991-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849387864338464768
author Matias Bermann
Alejandra Alvarez-Munera
Andres Legarra
Ignacio Aguilar
Ignacy Misztal
Daniela Lourenco
author_facet Matias Bermann
Alejandra Alvarez-Munera
Andres Legarra
Ignacio Aguilar
Ignacy Misztal
Daniela Lourenco
author_sort Matias Bermann
collection DOAJ
description Abstract Background Likelihood-based inferences such as variance components estimation and hypothesis testing need logarithms of the determinant (log-determinant) of high dimensional matrices. Calculating the log-determinant is memory and time-consuming, making it impossible to perform likelihood-based inferences for large datasets. Results We presented a method for approximating the log-determinant of positive semi-definite matrices based on repeated matrix–vector products and complex calculus. We tested the approximation of the log-determinant in beef and dairy cattle, chicken, and pig datasets including single and multiple-trait models. Average absolute relative differences between the approximated and exact log-determinant were around 10–3. The approximation was between 2 and 500 times faster than the exact calculation for medium and large matrices. We compared the restricted likelihood with (approximated) and without (exact) the approximation of the log-determinant for different values of heritability for a single-trait model. We also compared estimated variance components using exact expectation–maximization (EM) and average information (AI) REML algorithms, against two derivative-free approaches using the restricted likelihood calculated with the log-determinant approximation. The approximated and exact restricted likelihood showed maxima at the same heritability value. Derivative-free estimation of variance components with the approximated log-determinant converged to the same values as EM and AI-REML. The proposed approach is feasible to apply to any data size. Conclusions The method presented in this study allows to approximate the log-determinant of positive semi-definite matrices and, therefore, the likelihood for datasets of any size. This opens the possibility of performing likelihood-based inferences for large datasets in animal and plant breeding.
format Article
id doaj-art-a93ec9ef0e1c4803a83a3a20bf8c0797
institution Kabale University
issn 1297-9686
language deu
publishDate 2025-08-01
publisher BMC
record_format Article
series Genetics Selection Evolution
spelling doaj-art-a93ec9ef0e1c4803a83a3a20bf8c07972025-08-20T03:42:29ZdeuBMCGenetics Selection Evolution1297-96862025-08-0157111110.1186/s12711-025-00991-1Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative geneticsMatias Bermann0Alejandra Alvarez-Munera1Andres Legarra2Ignacio Aguilar3Ignacy Misztal4Daniela Lourenco5Department of Animal and Dairy Science, University of GeorgiaDepartment of Animal and Dairy Science, University of GeorgiaCDCBInstituto Nacional de Investigación Agropecuaria (INIA)Department of Animal and Dairy Science, University of GeorgiaDepartment of Animal and Dairy Science, University of GeorgiaAbstract Background Likelihood-based inferences such as variance components estimation and hypothesis testing need logarithms of the determinant (log-determinant) of high dimensional matrices. Calculating the log-determinant is memory and time-consuming, making it impossible to perform likelihood-based inferences for large datasets. Results We presented a method for approximating the log-determinant of positive semi-definite matrices based on repeated matrix–vector products and complex calculus. We tested the approximation of the log-determinant in beef and dairy cattle, chicken, and pig datasets including single and multiple-trait models. Average absolute relative differences between the approximated and exact log-determinant were around 10–3. The approximation was between 2 and 500 times faster than the exact calculation for medium and large matrices. We compared the restricted likelihood with (approximated) and without (exact) the approximation of the log-determinant for different values of heritability for a single-trait model. We also compared estimated variance components using exact expectation–maximization (EM) and average information (AI) REML algorithms, against two derivative-free approaches using the restricted likelihood calculated with the log-determinant approximation. The approximated and exact restricted likelihood showed maxima at the same heritability value. Derivative-free estimation of variance components with the approximated log-determinant converged to the same values as EM and AI-REML. The proposed approach is feasible to apply to any data size. Conclusions The method presented in this study allows to approximate the log-determinant of positive semi-definite matrices and, therefore, the likelihood for datasets of any size. This opens the possibility of performing likelihood-based inferences for large datasets in animal and plant breeding.https://doi.org/10.1186/s12711-025-00991-1
spellingShingle Matias Bermann
Alejandra Alvarez-Munera
Andres Legarra
Ignacio Aguilar
Ignacy Misztal
Daniela Lourenco
Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
Genetics Selection Evolution
title Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
title_full Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
title_fullStr Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
title_full_unstemmed Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
title_short Monte Carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
title_sort monte carlo approximation of the logarithm of the determinant of large matrices with applications for linear mixed models in quantitative genetics
url https://doi.org/10.1186/s12711-025-00991-1
work_keys_str_mv AT matiasbermann montecarloapproximationofthelogarithmofthedeterminantoflargematriceswithapplicationsforlinearmixedmodelsinquantitativegenetics
AT alejandraalvarezmunera montecarloapproximationofthelogarithmofthedeterminantoflargematriceswithapplicationsforlinearmixedmodelsinquantitativegenetics
AT andreslegarra montecarloapproximationofthelogarithmofthedeterminantoflargematriceswithapplicationsforlinearmixedmodelsinquantitativegenetics
AT ignacioaguilar montecarloapproximationofthelogarithmofthedeterminantoflargematriceswithapplicationsforlinearmixedmodelsinquantitativegenetics
AT ignacymisztal montecarloapproximationofthelogarithmofthedeterminantoflargematriceswithapplicationsforlinearmixedmodelsinquantitativegenetics
AT danielalourenco montecarloapproximationofthelogarithmofthedeterminantoflargematriceswithapplicationsforlinearmixedmodelsinquantitativegenetics