Gene editing to identify weeds
A proposal by Pedro M.P. Correia et al, from the University of Copenhagen, suggests that horticulture generally could make use of naturally occurring selection processes to improve crop yields.
Most high-yielding crops are susceptible to abiotic and biotic stresses, making them particularly vulnerable to the potential effects of climate change. A possible alternative is to accelerate the domestication of wild plants that are already tolerant to harsh conditions, and to increase their yields by methods such as gene editing. He sees, however, that crops’ wild progenitors could potentially compete with the newly domesticated plants, reducing yields. To improve the recognition of weeds, the proposal is to use gene editing techniques to introduce traits into these newly domesticated crops, allowing for visual recognition of the crops by weeding robots that have been trained by machine learning.
By changing the colour of newly domesticated plants, and by including genes associated with blueberries and carrots, weeding robots could be more easily trained to recognise the colour of newly domesticated plants from their still-wild cousins.