ASRgenomics: filling the gap on processing molecular data for quantitative genetics

ASRgenomics: filling the gap on processing molecular data for quantitative genetics

The VSNi Team

29 June 2022
image_blog

Optimizing your genomic workflow

Most breeding programs are supported by an array of genomic information that will provide new options to increase the rates of genetic gain. However, performing statistical analyses with molecular data can be a difficult task. This type of data has to communicate properly with available phenotypic and pedigree data. The overall success of this integration depends on a set of checks, verifications, filters, and careful preparation of all these datasets in order to be able to fit genetic models successfully and to obtain the required output to make correct decisions.

Essential questions for genomic analysis

The workflow of molecular data-driven analysis varies based on the source of the datasets and of course, on personal preferences. Nevertheless, regardless of these aspects, an efficient genomics pipeline should rely on answering some of the following questions:

  • How to filter out bad quality markers?
  • How to remove redundant marker information (e.g., pruning)?
  • How to check the genotypes sample for underlying population structure?
  • Which algorithms to use for generating a genomic relationship matrix (GRM)?
  • Is the quality and reliability of the GRM suitable for an analysis (i.e., are there duplicates or other inconsistencies)?
  • How to modify a GRM if there are duplicates or other inconsistencies?
  • How to eliminate bias in a GRM using pedigree information?
  • How to combine the GRM with the pedigree to obtain the hybrid matrix H used in ssGBLUP?
  • How to obtain a well-conditioned inverse of the GRM?
  • How to assess if the inverse of a GRM is good enough for genomics modeling?
  • How to efficiently subset and match my datasets (phenotypic, molecular, etc.)?

Introducing ASRgenomics

We developed ASRgenomics to help deal with the above questions. This is a free to use R library which can be downloaded from the ASReml knowledgbase. It is a compilation of proven routines developed over several years of study and hands-on experience in the field. ASRgenomics was built with advanced statistical modeling in mind and it fills a gap by helping you make sure your analyses are as efficient and accurate as they can be with several explicit diagnostic tools.

Enhancing genomic analyses

The package is aimed at geneticists and breeders with the purpose of improving their experience with genomic analyses, such as Genomic Selection (GS) and Genome Wide Association Studies (GWAS), in a straightforward and efficient manner. The main capabilities of the package include:

  • Preparing and exploring pedigree, phenotypic and genomic data.
  • Calculating and evaluating genomic matrices and their inverse.
  • Complementing and expanding results from genomic analyses.

Flexible functions for tailored workflows

The functions included within ASRgenomics are very flexible and can be used for a tailored workflow from raw molecular data to well-behaved model-ready matrices. Additionally, ASRgenomics is capable of seamlessly preparing genomic datasets for integration with ASReml-R to fit linear mixed models (LMMs; e.g., GBLUP or ssGBLUP).

Try ASRgenomics for free

Please try this free library and check out the user guide included withinin the doc folder inside the download package for a walk-though of the features along with details of the methods.