The Power of GWAS with ASReml-R

The Power of GWAS with ASReml-R

The VSNi Team

01 May 2024

Uncovering which genes are associated with specific traits is highly valuable for researchers and breeding programs. This allows the selection of favorable trait expression based on genomic information instead of using more time-consuming traditional evaluation methods.   

In brief, when using Genome-Wide Association Studies (GWAS), researchers are trying to pinpoint molecular markers that are correlated with changes in phenotypic expression. The basic concept is that, when a correlation is found with a marker, it is because its position in the DNA is close enough to the actual coding gene, so that they often segregate together during the reproduction process, and as such both are passed to the next generation in a specific allelic combination. In this case, the marker and the gene are said to be in linkage disequilibrium (LD), and for this reason, the marker gives us information about the gene. 

GWAS can test simultaneously large numbers of molecular markers distributed throughout the whole genome and identify which markers are associated with the trait under study (e.g., disease resistance, plant yield, or meat quality). This is a valuable tool, as breeders, based on this information, can select a set of individuals, using their genomic information; hence, these individuals can be identified early on and they do not necessarily require phenotypic assessment over their life cycle. 

GWAS methods have already proven their worth, unfortunately, their implementations are often limited by their inability to fully capture the nuances of data structures, particularly when it comes to considering several random and fixed effects simultaneously. This is where ASReml-R, our powerful mixed model software, and its companion R library ASRgwas come into play.

One of the key strengths of ASReml-R lies in its ability to accommodate a wide range of fixed and random effects as well as complex data structure. Whether dealing with replicated measurements, binary response variables, or even correlated structures, the software adapts to the needs of the researcher, rather than forcing them to conform to rigid methodological constraints.

Integrating ASReml-R with our library ASRgwas harnesses the strengths of both tools and empower researchers with unparalleled flexibility in GWAS modeling. It allows to tailor analyses to suit specific research objectives and data characteristics, enabling even the most complex analyses to be executed with precision and efficiency.

The ASRgwas library follows a three-step approach to GWAS analysis:

  • Data Preparation and Quality Control: ASRgwas ensures that the phenotypic and genotypic data are thoroughly evaluated and pre-processed. Notably, this step involves filtering out low-quality markers, imputing missing data, and calculating key matrices. 
  • Model Fitting: ASReml-R and ASRgwas provide the flexible framework for fitting complex model structures over the full set of markers. 
  • Post-GWAS Analysis and Validation: ASRgwas offers tools for displaying, interpreting, and validating the results, such as assessing marker significance, performing backward selection, and visualizing results through graphical outputs.

By using ASRgwas together with ASReml-R, researchers gain access to a comprehensive and flexible GWAS platform, empowering them to uncover the genetic underpinnings of complex traits with unprecedented precision, flexibility, and efficiency. (ASReml-R is a licensed product and ASRgwas is a free library but you need an ASReml-R license to use together.)

To learn more about harnessing the power of ASReml-R and the ASRgwas library for your GWAS analyses, we welcome and encourage you to watch this webinar ‘A flexible GWAS modelling workflow with ASReml-R’ hosted by Dr. Salvador Gezan.