fbpx

ASReml: a powerful tool in crop and animal breeding

C. Supakorn

 

The world’s increasing population is driving the demand to optimize the production of more reliable and higher quality crop and livestock products.

 

Agricultural science, including genomic research programs, has resulted in improved drought tolerance in corn, wheat, and other crops, and disease resistance in cattle, pig, poultry, and other livestock.

 

Today, new technologies in crop and animal breeding are making a significant contribution to accelerating the rate of genetic gain and transforming agriculture. For example, the development of linear mixed models, coupled with the availability of statistical software (such as ASReml) to apply this methodology, has resulted in a major paradigm shift.

 

ASReml is a comprehensive and powerful statistical software package for linear mixed model analysis used by animal and plant breeders, and agricultural, horticultural, aquaculture, environmental and medical scientists around the world.

 

ASReml arose out of a collaboration between Arthur Gilmour and Brian Cullis (New South Wales Department of Primary Industries, Australia) with Robin Thompson and Sue Welham (Rothamsted Research, United Kingdom) – scientists with world-leading expertise in the analysis of animal and crop breeding data and the development of linear mixed model methods that are both statistically and computationally efficient.

ASReml fits linear mixed models using Restricted Maximum Likelihood (REML). Its REML routine is fast and efficient – it uses the Average Information (AI) algorithm and sparse matrix methods – enabling users to analyze large data sets. Moreover, ASReml can analyze complex data, such as diallel and multivariate data, and accommodates pedigree and SNP information easily.

ASReml provides theoretically advanced approaches for the mixed model analysis of discrete, categorical, continuous, and non-normal response data, as well as spatial, repeated measures (longitudinal), multivariate, complex design, and unbalanced data sets.

 

VSNi offers ASReml as a standalone package or ASReml-R for use within the R environment. These are the powerful tools for fitting mixed models and are based on the same computational core. ASReml standalone represent a text file interface and wider workspace, making it useful for complex problems. ASReml-R is convenient for regular users of the R environment, and it is simpler to manipulate data. Both packages are compatible with high-performance computing in a cloud environment.

 

Take advantage of our powerful tool for mixed model analysis! Check out ASReml software website for more detail here.