BACKGROUND: Boiled yam key quality attributes typical for West African consumers are: crumbly, easy to break, and sweet taste. New yam varieties are being developed but high or medium throughput tools to assess the required quality traits and their range of acceptance are limited. This study assessed the acceptance thresholds of these quality attributes and established the predictive models for screening yam varieties that meet the required consumers' preferences. RESULTS:Overall liking was associated with sweet taste, crumbly and easy to break (r values 0.502, 0.291 and -0.087, respectively). These parameters and selected biophysical parameters highly discriminated the boiled yam varieties. Crumbly texture and easy to break were wellpredicted by penetration force and dry matter, whereas sweet taste by dry matter and sugar intensity. A high crumbliness and sweet taste are preferred (sensory scores above 6.19 and 6.22 for crumbly and sweet taste, respectively, on a 10 cm unstructured line scale), while a too high easiness to break is disliked (sensory scores ranging from 4.72 to 7.62). Desirable biophysical targets were between 5.1 and 7.1 N for penetration force, dry matter around 39% and sugar intensity below 3.62 g/100g. Some improved varieties fulfilled the acceptable thresholds, and the screening was improved through the deviation from optimum. CONCLUSION:The acceptance thresholds and the deviation from optimum for boiled yam assessed through the instrumental measurements are promising tools for yam breeders.

BACKGROUNDThe purpose of this study was to investigate the potential of hyperspectral imaging for the characterization of cooking quality parameters, dry matter content (DMC), water absorption (WAB), and texture in cassava genotypes contrasting for their cooking quality.RESULTSHyperspectral images were acquired on cooked and fresh intact longitudinal and transversal slices from 31 cassava genotypes harvested in March 2022 in Colombia. Different chemometric methods were tested for the quantification of DMC, WAB, and texture parameters. Data analysis was conducted through partial least squares regression, K nearest neighbors regression, support vector machine regression and CovSel multiple linear regression (CovSel_MLR). Efficient performances were obtained for DMC using CovSel_MLR with, coefficient of multiple determination , root‐mean‐square error of prediction RMSEP = 0.96 g/100 g, and ratio of the standard deviation values RPD = 3.60. High heterogeneity was observed between contrasting genotypes. The predicted distribution of DMC within the root can be homogeneous or heterogeneous depending on the genotype. Weak predictions were obtained for WAB and texture parameters.CONCLUSIONSThis study showed that hyperspectral imaging could be used as a high‐throughput phenotyping tool for the visualization of DMC in contrasting cooking quality genotypes. Further improvement of protocols and larger datasets are required for WAB and texture quality traits. © 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Wood characteristics of trees grown in agroforestry systems are little studied, even if growth conditions are different from conventional stands. This work aimed to determine the impact of the agroforestry system on the heartwood formation process of hybrid walnut (Juglans regia × nigra) trees, especially the resulting extractive contents. Ethanol and water extractions were successively performed on wood samples taken across the diameter of the trunk of agroforestry (AF) and forest (FC) walnut trees to get the radial distribution of the extractive contents. All the samples were analyzed by NIR-spectroscopy and NIR-hyperspectral imaging. Statistical discriminant models were developed to classify the samples from both different forestry systems, according to their chemical composition. The results indicated no significant differences between the values of extractive contents of AF and FC walnut woods, whatever the radial position. At the intra-tree scale, the quantity of extractives does not increase significantly with the radial position. However, partial least squares-discriminant analysis (PLS-DA) regression models, developed with NIRS measurements, showed that significant chemical differences exist between AF and FC trees, especially for extractives composition and lignin content. This allowed to classify wood specimens from both forestry systems. These results were confirmed by hyperspectral camera analyses.