Fig. 5: Analysis of multi-year agronomic performance of AKT2var and EV control plants in field trials.
From: Engineering vascular potassium transport increases yield and drought resilience of cassava

a, Pearson correlation of agronomic traits per year using data from both AKT2var and EV plants. Positive and negative correlations are shown in red and blue, respectively. Significance was tested using a two-sided Student’s t-test. TFW, total fresh weight; DMC, dry matter content. b, Line performance for each year. Spatially corrected data values of replicates (at least ten) are shown for years and traits. c, Line performance after spatial and temporal correction of the raw data (Methods). Adjusted means (BLUEs) ± s.e. of genotype effects for each line and trait are shown. The values represent BLUEs obtained from a linear mixed-effects model of spatially corrected trait values, with line fitted as a fixed effect and year and replicate as random effects (trait spatially corrected ~ line + (1|year) + (1|replicate)). The error bars show the model-derived standard errors of the adjusted means. In each plot, the dashed horizontal line indicates the mean value of the reference line, and the shaded grey area represents its mean ± s.e. The asterisks denote significance levels as follows: not significant (NS, P > 0.05), significant (*P < 0.05), highly significant (**P < 0.01) and very highly significant (***P < 0.001).