Scribd Logo
Upload

Welcome to Scribd!

  • 30 views

    WOMBAT: A Tool For Estimation of Genetic Parameters - Highlights and Updates

    Uploaded by

    Yhan Carlos Rojas de la Crúz
    WOMBAT is a software package for estimating genetic parameters using linear mixed models. It can handle large datasets with millions of equations. Recent updates include support for very large datasets using 64-bit integers and faster processing of large analyses. WOMBAT allows for many model types including fixed effects, genetic effects, maternal effects, and more. It provides functions for estimating variance components, predicting breeding values, and related analyses. The software is free to download and supports both Linux and Windows platforms.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    WOMBAT: A Tool For Estimation of Genetic Parameters - Highlights and Updates

    Uploaded by

    Yhan Carlos Rojas de la Crúz
    30 views1 page
    WOMBAT is a software package for estimating genetic parameters using linear mixed models. It can handle large datasets with millions of equations. Recent updates include support for very large datasets using 64-bit integers and faster processing of large analyses. WOMBAT allows for many model types including fixed effects, genetic effects, maternal effects, and more. It provides functions for estimating variance components, predicting breeding values, and related analyses. The software is free to download and supports both Linux and Windows platforms.

    Original Title

    icqg6

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    WOMBAT is a software package for estimating genetic parameters using linear mixed models. It can handle large datasets with millions of equations. Recent updates include support for very large datasets using 64-bit integers and faster processing of large analyses. WOMBAT allows for many model types including fixed effects, genetic effects, maternal effects, and more. It provides functions for estimating variance components, predicting breeding values, and related analyses. The software is free to download and supports both Linux and Windows platforms.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    30 views1 page

    WOMBAT: A Tool For Estimation of Genetic Parameters - Highlights and Updates

    Uploaded by

    Yhan Carlos Rojas de la Crúz
    WOMBAT is a software package for estimating genetic parameters using linear mixed models. It can handle large datasets with millions of equations. Recent updates include support for very large datasets using 64-bit integers and faster processing of large analyses. WOMBAT allows for many model types including fixed effects, genetic effects, maternal effects, and more. It provides functions for estimating variance components, predicting breeding values, and related analyses. The software is free to download and supports both Linux and Windows platforms.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 1

    WOMBAT: A tool for estimation of genetic parameters –

    highlights and updates


    Karin Meyer
    Animal Genetics and Breeding Unit, University of New England, Armidale, Australia

    Purpose Main features


    WOMBAT is a software package for linear WOMBAT accommodates many different models & analyses and has multiple options for fine tuning
    mixed model analysis in quantitative genetics WOMBAT can handle large analyses with millions of equations
    Focus on REML estimation of (co)var- Analyses Models Options
    iance components
    • Uni- or multivariate REML estimation of Specify via parameter file: Specify via command line or parameter file:
    Targeted at animal breeding problems but variance components & genetic parameters • Fixed effect and covariables • Ordering of mixed model equations
    suitable for many related applications
    – Full (default) or reduced rank estimates of • Genetic effects • Strategy to locate maximum of likelihood
    Continuous traits covariance matrices – choice of relationship matrix – Algorithm, iterates, convergence criterion
    – Penalised likelihood estimation to reduce
    • Maternal or paternal effects • Parameterisation
    sampling variance
    • Indirect (“social”) genetic effects • Matrix storage: sparse or dense
    Recent changes & additions • Prediction only (BLUP)
    • Additional random effects • Numerical settings, e.g. operational zeros
    – Direct solution • Genetic groups • Pedigree checks & manipulation
    • Version using long (8 byte) Integer variables
    – Iterative solution (very large problems)
    – allows sparse storage of matrices with • Random regression • Invoke auxiliary modules
    – Special option for single step models
    > 2.14 billion elements • Factor-analytic models
    • Options to override default program limits to
    suit individual analysis & RAM available
    • Improved set-up steps for large analyses
    incl. faster build of the NRM inverse Auxiliary modules Get it!
    • Implementation of GWAS via backsolving
    for marker effects from breeding values • Inversion of positive (semi-) definite matrices: sparse or dense mode I Free download from WOMBAT home page
    • ‘Pooling’ estimates of covariance components from analyses of subsets of traits http://didgeridoo.une.edu.au/km/wombat.php
    – make matrices positive definite < Executable programs
    • Simulation of records for given variances, data & pedigree structure ± Linux (multi-threaded)
    Technical Details q Windows
    • Write out mixed model equation
     User Manual
    • ‘Super-nodal’ factorisation, auto-differentia- • Sampling based approximation of standard errors for variance components
    tion & inversion of sparse coefficient matrix  PDF file (A4)
    • GWAS (EMMAX type) ü Online version
    • Written in Fortran • H−1 & friends: Calculations of relationship matrices utilising genomic information
    ü Frequently Asked Questions
    – Compiled using ifort or gfortran; 64 bit −1
    – G, G , A22 , A−1 −1
    H , log |H|, Diag(H)
    22 ,
    – Parallelisation via OpenMP & Intel MKL I Suite of worked examples
    – A with Meta-founders, H−1 with unknown parent groups
    −1
    routines (Linux version)
    Click on links in bold to download

    kmeyer@une.edu.au 6th International Conference of Quantitative Genetics, Virtual, November 3–13, 2020

    You might also like