The transcriptome landscape of wheat and related genomic resources
The recent improvements in the assembly of the hexaploid wheat genome have the potential to accelerate the pace of genetic research in this important crop. At the same time, the amount and complexity of the data can feel overwhelming given its polyploid and large 16 Gb genome. Ricardo Ramírez-González is going to present visualisations to explore the relationship between homoeologous genes that intuitively represent the expression bias across the transcriptome. His lab analysed 850 public RNA-Seq expression samples across different varieties, tissues, ages and under different environmental conditions.
Ternary plots are used to get an overview of the whole transcriptome or for a single set of homoeologous genes (triad) across different tissues. The distance between the expression biases across tissues is used to categorise the expression of each triad as stable or dynamic. This visualisation technique helps to answer the longstanding question of dosage effects in polyploid organisms and characterises early steps towards sub-functionalisation. Including the genomic coordinates and predicted gene ontologies, Ramírez-González's group discovered that genes located in the most distal regions of the chromosomes and related to environmental response tend to be more dynamic. Likewise, housekeeping genes are located in proximal regions and show more stable expression across genomes. Using this novel metric, they discovered that dynamic genes are under more relaxed selection pressure than stable genes, linking variation in relative expression to sequence variation in coding regions.
To make the expression data more accessible to researchers and breeders in the wheat community, Ramírez-González's group developed expVIP, an expression browser. The interactive visualisation allows users to group the different experiments in high-level, intermediate, and fine-grained factors. Expression data can be represented for a group of genes, a single gene, or for all gene homoeologs. These features allow interactive hypothesis testing and have been used to narrow down candidate genes inside a target genetic interval.