EMBO Press is an editorially independent publishing platform for the development of EMBO scientific publications.

Physiological measurements of nitrate (NO₃⁻) uptake by roots have defined two systems of high and low affinity uptake. In Arabidopsis, genes encoding both of these two uptake systems have been identified. Most is known about the high affinity transport system (HATS) and its regulation and yet measurements of soil NO₃⁻ show that it is more often available in the low affinity range above 1 mM concentration. Several different regulatory mechanisms have been identified for AtNRT2.1, one of the membrane transporters encoding HATS; these include feedback regulation of expression, a second component protein requirement for membrane targeting and phosphorylation, possibly leading to degradation of the protein. These various changes in the protein may be important for a second function in sensing NO₃⁻ availability at the surface of the root. Another transporter protein, AtNRT1.1 also has a role in NO₃⁻ sensing that, like AtNRT2.1, is independent of their transport function. From the range of concentrations present in the soil it is proposed that the NO₃⁻-inducible part of HATS functions chiefly as a sensor for root NO₃⁻ availability. Two other key NO₃⁻ transport steps for efficient nitrogen use by crops, efflux across membranes and vacuolar storage and remobilization, are discussed. Genes encoding vacuolar transporters have been isolated and these are important for manipulating storage pools in crops, but the efflux system is yet to be identified. Consideration is given to how well our molecular and physiological knowledge can be integrated as well to some key questions and opportunities for the future.

Major increases in crop yield are required to keep pace with population growth and climate change. Improvements to the architecture of crop roots promise to deliver increases in water and nutrient use efficiency but profiling the root phenome (i.e. its structure and function) represents a major bottleneck. We describe how advances in imaging and sensor technologies are making root phenomic studies possible. However, methodological advances in acquisition, handling and processing of the resulting 'big-data' is becoming increasingly important. Advances in automated image analysis approaches such as Deep Learning promise to transform the root phenotyping landscape. Collectively, these innovations are helping drive the selection of the next-generation of crops to deliver real world impact for ongoing global food security efforts.

In plant phenotyping, it has become important to be able to measure many features on large image sets in order to aid genetic discovery. The size of the datasets, now often captured robotically, often precludes manual inspection, hence the motivation for finding a fully automated approach. Deep learning is an emerging field that promises unparalleled results on many data analysis problems. Building on artificial neural networks, deep approaches have many more hidden layers in the network, and hence have greater discriminative and predictive power. We demonstrate the use of such approaches as part of a plant phenotyping pipeline. We show the success offered by such techniques when applied to the challenging problem of image-based plant phenotyping and demonstrate state-of-the-art results (>97% accuracy) for root and shoot feature identification and localization. We use fully automated trait identification using deep learning to identify quantitative trait loci in root architecture datasets. The majority (12 out of 14) of manually identified quantitative trait loci were also discovered using our automated approach based on deep learning detection to locate plant features. We have shown deep learning–based phenotyping to have very good detection and localization accuracy in validation and testing image sets. We have shown that such features can be used to derive meaningful biological traits, which in turn can be used in quantitative trait loci discovery pipelines. This process can be completely automated. We predict a paradigm shift in image-based phenotyping bought about by such deep learning approaches, given sufficient training sets.