Back

Gene-metabolite networks reveal physiological trade-offs but not transcriptional co-regulation between carotenoids and dry matter in cassava (Manihot esculenta) roots

Villwock, S. S. C.; Gomez, K. M.; Fish, T.; Lee, J.; Doherty, A.; White, A.; Thannhauser, T. W.; Gore, M. A.; Jannink, J.-L.

2026-07-17 plant biology
10.64898/2026.07.11.737982 bioRxiv
Show abstract

Provitamin A biofortification of cassava (Manihot esculenta) is constrained by a negative genetic correlation between total carotenoid (TC) and dry matter (DM) contents, but its underlying biological mechanism remains unclear. We examined this relationship in 24 African, Latin American, and hybrid cassava genotypes by measuring gene expression and metabolites in inner and outer storage root layers that varied spatially in carotenoid accumulation, enabling comparisons across and within genotypes. TC, and particularly upstream cis-carotenes, were negatively associated with DM after accounting for genotype, and this relationship strengthened across developmental timepoints, consistent with a physiological component of the trade-off. Gene-metabolite co-expression networks constructed from genetic, non-genetic, and overall phenotypic trait components showed no significant topological overlap between carotenoid- and starch-related gene sets, indicating that the TC/DM relationship is not driven by transcriptional co-regulation of the main pathways, although individual associations between carotenoids and carbohydrate-related genes suggested possible indirect metabolic links. Instead, TC accumulation was most strongly associated with regulatory, stress-, redox-, and plastid-related genes, enriched for heme-binding and oxidoreductase functions. These findings suggest that the TC/DM trade-off is not a fixed constraint, and identify candidate regulators of carotenoid content for future functional validation to support cassava provitamin A biofortification. HighlightGenetic, spatial, and developmental variation in cassava storage roots implicate cis-carotenes, but not direct transcriptional regulation of the carotenoid and starch pathways, in the trade-off between carotenoids and dry matter.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.