Conclusion
We propose two main root phene-based strategies to accelerate the development of new cultivars better adapted to low-input environments in Africa.
The first is to identify simple phenes that have broadest positive influence on enhanced performance and that also minimize trade-offs. Towards this, long and dense root hairs are likely selection criteria as greater root hair length and density promotes exudation and have positive effects on the beneficial microbial activity within the rhizosphere (Rongsawat, Peltier, Boyer, Véry & Sentenac 2021). Additionally, root plasticity phenes could be another potential selection criteria as for topsoil root branching plasticity could be beneficial upon partial dry-down (for rice in AWD agroecosystems) while subsoil root plasticity could be beneficial during prolonged drought stress particularly in the reproduction and grain filling stage (for sorghum and pearl millet in arid and semi-arid agroecosystems).
The second major strategy is to understand and target phene synergisms and integrated phenotypes. Synergisms between root phenes are defined as interactions that have more than additive effect, as in the case of long and dense root hairs paired with shallow root system architecture for P acquisition (Miguel et al. 2015). Integrated phenotypes would clearly affect the utility of selecting for a single component phene without selecting for their complementary phenotypes. For example, the utility of high conductance capacity xylem likely depends on root phenes that affect rooting depth since deep roots can access and thus transport greater volumes of soil water (Strock, Burridge, Niemiec, Brown & Lynch 2020). The development of root structural and functional models for crops such as sorghum or pearl millet that can evaluate the effects of architectural and anatomical phenes in changing soil environments and its effects on root functions will be particularly useful (Ndour, Pradal & Lucas 2017a). Other less well characterized phene assemblages, especially those involving transpiration, should be further investigated and validated in particular stress scenarios (Strock et al. 2019; Klein, Schneider, Perkins, Brown & Lynch 2020). Considering resource acquisition and use, especially that of water within the context of phenology, leads to acknowledging the importance of interactions among roots and shoots for timely water use across the crop cycle (Vadez, Kholova, Medina, Kakkera & Anderberg 2014). In that regard, combining root models with crop models could potentially link above-ground phenes to root phenes, the former serving as a proxy for root function (Benes et al. 2020).
In order to make these innovations readily available to breeders, researchers and breeders need to work together to validate phene utility, develop phenotyping protocols including field sites, the type of genetic material to work with (RILs, NILs, tester lines, germplasm deployment strategy) and ultimately identify marker or genes controlling beneficial phenes. As in the case of common bean presented in case study 1, identifying a selection strategy, including the type of phene for the appropriate stage, would be particularly useful. To maximize deployment of improved cultivars and to then secure the adoption of those improved cultivars, social scientists and farmers should be integrated in the selection process (Ameleworket al. 2016). The inclusion of useful root phenes in such approaches may help to stimulate a sustainable Green Revolution in Africa.