This study investigated the trait preferences for cassava in the context of climate change and conflict stressors among value-chain actors in Nigeria to strengthen social inclusion and the community-resilience outcomes from breeding programs. Multi-stage sampling procedures were used to select and interview male and female value-chain participants in the Osun, Benue and Abia States. The results indicated that farmers preferred cassava traits such as drought tolerance, early bulking, multiple-product use and in-ground storability to strengthen resilience. Climate change and challenges related to social change shaped the response strategies from both genders, and influenced trait preferences, including the early re-emergence of cassava leaves, stems that had ratooning potential, and especially the root milking that was important among female respondents. The major response strategies employed by men included frequent farm visits to prevent theft and engaging in non-agricultural livelihoods. Those employed by women included backyard farming, early harvesting, having preferences for food with fewer processing steps, and depending on remittances. The resilience capacity was higher for men than for women due to their better access to assets, as well as their abilities to relocate their farms and out-migrate in search of other livelihoods. Considering gendered cassava traits, and enhancing their resilience and response strategies, can complement efforts to make breeding more socially inclusive, resilient, and anticipatory to future challenges created by climate and related social changes.

Experimental error, especially through genotype misclassification and pedigree errors, negatively affects breeding decisions by creating ‘noise’ that compounds the genetic signals for selection. Unlike genotype-by-environment interactions, for which different methods have been proposed to address, the effect of ‘noise’ due to pedigree errors and misclassification has not received much attention in most crops. We used two case studies in sweetpotato, based on data from the International Potato Center’s breeding program to estimate the level of phenotype misclassification and pedigree error and to demonstrate the consequences of such errors when combining phenotypes with the respective genotypes. In the first case study, 27.7% phenotype misclassification was observed when moving genotypes from a diversity panel through in-vitro , screenhouse and field trialing. Additionally, 22.7% pedigree error was observed from misclassification between and within families. The second case study involving multi-environment testing of a full-sib population and quantitative trait loci (QTL) mapping showed reduced genetic correlations among pairs of environments in mega-environments with higher phenotype misclassification errors when compared to the mega-environments with lower phenotype misclassification errors. Additionally, no QTL could be identified in the low genetic correlation mega-environments. Simulation analysis indicated that phenotype misclassification was more detrimental to QTL detection when compared to missingness in data. The current information is important to inform current and future breeding activities involving genomic-assisted breeding decisions in sweetpotato, and to facilitate putting in place improved workflows that minimize phenotype misclassification and pedigree errors.

Soybean is a major legume crop in Togo. However, there is no scientific basis for guiding planting density recommendations to enhance yield. This study evaluated six soybean varieties across multiple planting densities in two major soybean-growing areas of Togo over two years. Data were collected on morphological traits and yield and were subjected to variance analysis, heritability estimate and mean comparison using R software version 4.4.1. Heritability was high (>0.8) for most traits except for number of branches per plant (0.54) and pod weight per plant (0.50). The local check variety (TGX1910) consistently outperformed introduced lines at all planting densities, except for hundred-seed weight. Plant height increased with density, whereas branch number, pods per plant, and grains per plant decreased, reflecting soybean’s self-regulation. Grain yield ranged from 1.17 t/ha at 200,000 plants/ha to 3.83 t/ha at 333,000 plants/ha. Despite reduced individual plant productivity at higher densities, plot yield increased up to an optimal density of ~333,000 plants/ha, before declining at the highest density of 450,000 plants/ha. Wider row spacing (60 cm) slightly enhanced yield compared to narrower spacing (40 cm). These results demonstrate that optimal planting density maximizes population-level yield, and planting soybean at 333,000 plants/ha with 60 cm row spacing is recommended for Togo.