Improving Genomic Prediction in Cassava Field Experiments by Accounting for Interplot Competition

Abstract

Plants competing for available resources is an unavoidable phenomenon in a field. We conducted studies in cassava (Manihot esculenta Crantz) in order to understand the pattern of this competition. Taking into account the competitive ability of genotypes while selecting parents for breeding advancement or commercialization can be very useful. We assumed that competition could occur at two levels: (i) the genotypic level, which we call interclonal, and (ii) the plot level irrespective of the type of genotype, which we call interplot competition or competition error. Modification in incidence matrices was applied in order to relate neighboring genotype/plot to the performance of a target genotype/plot with respect to its competitive ability. This was added into a genomic selection (GS) model to simultaneously predict the direct and competitive ability of a genotype. Predictability of the models was tested through a 10-fold cross-validation method repeated five times. The best model was chosen as the one with the lowest prediction root mean squared error (pRMSE) compared to that of the base model having no competitive component. Results from our real data studies indicated that <10% increase in accuracy was achieved with GS-interclonal competition model, but this value reached up to 25% with a GS-competition error model. We also found that the competitive influence of a cassava clone is not just limited to the adjacent neighbors but spreads beyond them. Through simulations, we found that a 26% increase of accuracy in estimating trait genotypic effect can be achieved even in the presence of high competitive variance.

Genomic prediction of yield and root development in wheat under changing water availability

“…The increase in ACC for GY and NC were consistent with the marked change in the VCs pattern when including neighbor effects. Improvement of ACC when modeling neighbor effects were also found in the previously mentioned study for cassava where up to 25% increase in accuracy was observed [40]. The ACC obtained in this experiment was expected to be relatively low due to the small size of the population used to train the models.…”

“…For both GY and NC, the RVCs accounted by genomic effects (g) increased considerably but the line effects (l) decreased in the same amount. This indicates that if neighbor effects are excluded from the model a large part of these effects are picked up by the effects of l. In a study for cassava data [40], neighbor effects were investigated as competitive ability. From their simulation study, accuracy in estimating genotypic effects increased when including competitive effects in the model.…”

“…From their simulation study, accuracy in estimating genotypic effects increased when including competitive effects in the model. Through their analysis for nine trials and four traits, significant competition effects were observed for 12 of the 36 combinations [40]. In plants, genotype response to a studied specific stress can be complicated by competition effects from neighbors [30].…”

Interdisciplinary strategies to enable data-driven plant breeding in a changing climate

“…57 Much subsequent research has focused on the application of genomic prediction to breeding populations [58][59][60] and the elaboration of more flexible models that also include GEIs. [61][62][63][64][65] These models are, however, fundamentally simple, modeling the genotype-phenotype relationship as the sum of additive genetic effects. Increasing complexity by increasing the number of markers, accounting for epistasis, allowing GEIs, and considering non-linear relationships is a challenging statistical and computational problem.…”

The tricot citizen science approach applied to on-farm variety evaluation: methodological progress and perspectives

“…For example, for potato competition effects between plots consisting of single ridges seems unimportant for yield, as stolons rarely extend beyond ridges (Connolly et al, 1993). For cassava, interplot competition effects have been found to extend beyond the first row (Elias et al, 2018). For sweetpotato, interplot interaction is thought to be substantial due to above-ground competition (Grüneberg et al, 2019).…”