systemsGWAS is an integrative genome-wide association framework designed to identify the genetic architecture underlying observed phenotypes, latent biological traits, and systems-level interactions across genetics, environments, management, and multi-omic data.
This repository was originally developed as multiomicGWAS. The software is currently transitioning to systemsGWAS to reflect its expanded scope beyond conventional multi-omic GWAS.
Repository: https://github.com/bodeolukolu/multiomicGWAS
The repository name will be updated to systemsGWAS following completion of the accompanying methodological manuscripts.
Genome-wide association studies have traditionally focused on identifying loci controlling individual phenotypic traits. While highly successful, modern breeding increasingly requires understanding coordinated biological processes involving multiple interacting traits, environments, management practices, and host-associated omics.
systemsGWAS extends conventional GWAS by providing a unified framework capable of analysing:
- Conventional phenotypic traits
- Multi-trait analyses
- Secondary traits
- Host-associated microbiome and metagenome profiles
- Latent biological phenotypes
- Environment-specific trait architectures
- Future systems-level phenotypes such as the Developmental Blueprint Index (DBI)
The framework leverages GWASpoly for association analysis across ploidy levels (2x–8x) while providing automated data integration, relationship matrix construction, and flexible phenotypic modelling.
Relationship matrices are generated automatically, including:
- Genomic relationship matrices (GRM) using AGHmatrix
- Microbiome/metagenome kernels using Aitchison compositional distances
- Additional user-defined kernels (planned)
This architecture enables integrative systems-genetics analyses rather than treating genomic and omic datasets independently.
✔ GWAS across diploid through octoploid populations
✔ Automated genotype quality control
✔ Multiple GWASpoly genetic models
✔ Automatic genomic relationship matrix construction
✔ Integration of microbiome/metagenome abundance data
✔ Microbial taxa analysed as either:
- covariates
- independent phenotypes
✔ Automatic microbiome kernel construction
✔ Flexible phenotype processing
✔ Publication-quality figures
The long-term vision is to evolve systemsGWAS into a comprehensive systems genetics platform.
Planned capabilities include:
- Developmental Blueprint Index (DBI)
- user-defined latent phenotypes
- PCA-derived traits
- factor-analysis traits
- environment-specific GWAS
- stability GWAS
- plasticity GWAS
- G×E association analyses
Integration of
- genomic data
- transcriptomics
- metabolomics
- microbiome
- environmental variables
- management variables
within a unified association framework.
Future releases will support integration with crop systems modelling, allowing latent developmental phenotypes and environment-specific ideotypes to be analysed alongside genomic data.
| Version | Focus |
|---|---|
| v1.x | multiomicGWAS |
| v2.x | systemsGWAS architecture |
| v2.x | latent phenotype framework |
| v2.x | multi-environment GWAS |
| v3.x | systems genetics |
| v4.x | crop systems integration (APSIM support) |
Clone the repository
git clone https://github.com/bodeolukolu/multiomicGWAS.gitAfter the repository rename:
git clone https://github.com/bodeolukolu/systemsGWAS.gitDownload
run_parameters_systemsGWAS.R
Edit the analysis parameters and execute the pipeline.
The parameter file controls:
- genotype input
- phenotype input
- covariates
- microbiome/metagenome analyses
- GWAS models
- filtering
- output directories
- plotting options
Core packages include
- GWASpoly
- AGHmatrix
- sommer
- compositions
- mice
- qvalue
Data manipulation
- data.table
- dplyr
- stringr
- reshape2
- zoo
Visualisation
- ggplot2
- qqplotr
- heatmaply
- GGally
Statistics
- ppcor
Additional packages may be required for optional modules.
Applications of systemsGWAS include:
(To be updated following publication.)
If you use systemsGWAS in your research, please cite the accompanying publication once available.
Until then, please reference the GitHub repository:
https://github.com/bodeolukolu/multiomicGWAS
Bug reports, feature requests, and pull requests are welcome.
Suggestions for extending systemsGWAS to additional omic data types, latent phenotype analyses, or systems genetics applications are encouraged.
Dr. Bode A. Olukolu
Science Leader – Crop Genetics
Department of Primary Industries
Queensland, Australia
Email: bolukolu@utk.edu
This project follows Semantic Versioning:
Apache License 2.0
https://github.com/bodeolukolu/multiomicGWAS/blob/master/LICENSE