Although canned pulses are increasingly consumed for their convenience, texture - particularly firmness - remains a key factor in consumer acceptance. This study proposes a rapid and simple methodology for establishing acceptability thresholds for the firmness of canned chickpeas and lentils. Fifteen batches of chickpeas and lentils, from different cultivars, harvested in 2022 and 2023, were soaked and then canned according to a standardized protocol. Firmness was evaluated using trained sensory panels and with a mini Kramer shear cell attached to a texture analyzer. There was a high correlation between instrumental and sensory firmness (R² = 0.88 for chickpeas and R² = 0.80 for lentils), confirming the predictive value of instrumental measurements. By combining instrumental data with Just-About-Right testing, firmness acceptability thresholds were identified: 10.5–13.4 N/g of fresh weight (FW) for chickpeas and 9.4–12.4 N/g of FW for lentils. This low-cost, rapid approach provides a practical tool for quality control and process optimization in the production of canned pulses.

BACKGROUND Determinants of culinary qualities of East African highland cooking bananas (EAHCB) are not well known. This constrains the inclusion of user-preferred traits in breeding. The aim of this research was to quantify key indicators of user-preferred characteristics to enable selection of acceptable hybrids. RESULTS Qualitative characteristics that drive preference were big bunches (15-34 Kg), long straight/slightly curved fingers (12-23 cm), yellowness and soft texture. Descriptive sensory analysis of the intensity of colour and texture the 23 genotypes revealed that landraces Kibuzi, Mbwazirume, Nakitembe and Mpologoma had higher intensity of yellowness and lower intensity of hardness (softer) and a low score (≤ 1.0 ) of astringency taste. A preference test showed that they had higher acceptability scores. Biochemical, instrumental, and sensory data revealed correlations between sensory firmness and instrumental hardness (r=0.5), sensory firmness and amylopectin (r=-0.54), suggesting that qualitative descriptions can be predicted by instrumental and biochemical indicators. Significant (p<0.05) variations in amylose and total starch content were observed in different varieties. Moderate correlations between instrumental hardness and firmness in mouth (r=0.55), cohesiveness and firmness in the mouth (r=0.57), adhesiveness and firmness in the mouth (r=0.64) were observed. Surprisingly, carotenoids content was not correlated with yellowness in cooked matooke. However, positive correlations were observed between chroma (b*) parameters of raw matooke and sensorial assessed color on cooked samples. CONCLUSIONS Qualitative characteristis; the bunch, pulp colour and texture; that drive users-preference in the EAHCB were quantified, paving way for breeders to use them to select genotypes with these attributes early in the breeding process. This article is protected by copyright. All rights reserved.

Summary The quality of processed products can be adversely affected by uncontrollable batches of mangoes, which exhibit heterogeneous characteristics. This study aimed to establish predictive models for sugar and organic acid contents (dependent variables) in processed products using the initial compositions of fresh mangoes. Three mango cultivars (cv. ‘Kent’, cv. ‘Keo Romeat’, and cv. ‘Keo Chen’) were classified as low‐density and high‐density groups. Each group of mangoes at the green‐mature, mid‐ripe, and ripe stages was processed into pasteurised purees, dried slices, and mango chips. Prediction models were established using a mix of simple linear regression (SLR) based on the initial content and Analysis of Variance (ANOVA) to identify the impact of qualitative variables (ripening stage, cultivar‐density, and processing technique). In processed mangoes, 13% sucrose content was estimated to accumulate with the three qualitative variables, whereas glucose and fructose contents decreased from their initial levels by 10% and 7%, respectively. Processing techniques can predict the ratio of sugars/acids (S/A) in processed products, regardless of the ripening stage or cultivar‐density. Similar to S/A, citric acid and malic acid contents in mango products were significantly increased by processing techniques. The initial content and processes were insufficient to predict the final contents of the same parameters in processed mangoes; therefore, some models need to include the effects of ripening stage and cultivar‐density to improve the prediction. These relevant explanatory variables contributed significantly to the development of the models, resulting the accuracy of predictive models with normalised root mean square errors (NRMSEs) lower than 10%, except for malic acid (14.04%). In conclusion, it is feasible to estimate the sugar and acidity levels in processed mangoes, offering promising possibilities for ensuring consistent quality of mango‐based products.

Background: Cassava breeding research programs focused initially on agronomic performance but in recent years they have considered the processability and the organoleptic properties of the final product, to increase the adoption of new varieties. One important cassava foodstuff is attieke, a fermented and granular product, so it is necessary to determine criteria used by attieke processors to select raw cassava and the characteristics leading consumers to like this product. This study assessed the attieke production process, the criteria associated with the quality of attieke, the sensory drivers of consumer acceptance, and their thresholds.Results: The total processing yield of attieke varied according to the cassava variety and depended primarily on the fermentation-pressing yield. However, it was not correlated either with the peeling yield or with morphological characteristics of cassava roots. The production of a ton of attieke required about 150 h. Dry matter, organic acids, soluble sugars, total pectin, and the pH of raw material and attieke varied depending on the cassava variety. Ten discriminating sensory attributes of attieke were identified. Consumer testing showed that overall liking for attieke was associated with sourness, texture, and brightness. Acceptable sensory score thresholds were 1.67-2.18 for sour odor, 4.75 to 6.3 for cohesiveness, and 5.4 to 6.3 for 'mouthfeel sensation'. Attieke dry matter correlated positively with cohesiveness and moldability.Conclusions: Several potential solutions are discussed to improve the adoption of cassava varieties for attieke production. However, further studies need to be carried out to translate the sensory thresholds of texture attributes into robust instrumental methods because texture is an important attribute of attieke in addition to sourness.(c) 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

BACKGROUNDConsumers of boiled cassava in Africa, Latin America and Asia use specific preference criteria to evaluate its cooking quality, in terms of texture, colour and taste. To improve adoption rates of improved cassava varieties intended for consumption after boiling, these preference criteria need to be determined, quantified and integrated as post-harvest quality traits in the target product profile of boiled cassava, so that breeding programs may screen candidate varieties based on both agronomic traits and consumer preference traits.RESULTSSurveys of various end-user groups identified seven priority quality attributes of boiled cassava covering root preparation, visual aspect, taste and texture. Three populations of contrasted cassava genotypes, from good-cooking to bad-cooking, in three countries (Uganda, Benin, Colombia) were then characterized according to these quality attributes by sensory quantitative descriptive analysis (QDA) and by standard instrumental methods. Consumers' preferences of the texture attributes mealiness and hardness were also determined. By analysis of correlations, the consumers' preferences scores were translated into thresholds of acceptability in terms of QDA scores, then in terms of instrumental measurements (water absorption during boiling and texture analysis). The thresholds of acceptability were used to identify among the Colombian and Benin populations promising genotypes for boiled cassava quality.CONCLUSIONThis work demonstrates the steps of determining priority quality attributes for boiled cassava and establishing their corresponding quantitative thresholds of acceptability. The information can then be included in boiled cassava target product profiles used by cassava breeders, for better selection and adoption rates of new varieties. (c) 2024 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

BACKGROUNDPreviously, a lexicon and protocol for quantitative descriptive analysis (QDA) was established for the Uganda sweetpotato breeding program. The implication of QDA scores for priority sensory attributes on consumer preference should be determined to interpret results efficiently and make decisions effectively. The present study aimed to develop a gender-responsive decision tree to obtain an overall sweetpotato eating quality score to facilitate demand-led targeted breeding selection. It focused on Kamuli and Hoima districts (Uganda) and uses pre-lease advanced clones ('NKB3', 'NKB105', 'NKB135', 'D11' and 'D20'), released varieties ('NASPOT 8' and 'NAROSPOT 1') and landraces ('Muwulu-Aduduma', 'Umbrella'). RESULTSIncluding boiled sweetpotato sensory characteristics, namely mealy, sweet taste, sweetpotato smell, firm and not fibrous, in breeding design would benefit end-users, especially women given their role in varietal selection, food preparation and marketing. 'D20', 'NASPOT 8' and 'NAROSPOT 1' were most liked in both districts. 'NKB3' and 'D11' were the least liked in Hoima, whereas 'Muwulu-Aduduma' was the least liked in Kamuli. There was a positive correlation between color and overall liking (r(2) = 0.8) and consumers liked the color (average rating & GE; 6 on a nine-point hedonic scale) of all genotypes. Threshold values (average rating on 11-point scales) for consumer acceptability were identified (sweet taste = 6, sweetpotato aroma and flavor = 6, firmness = 3, and mealiness = 4). A regression decision tree tool was created to calculate an eating quality selection index when screening lines in breeding programs using the values. CONCLUSIONDecision trees that include consumer needs and gender considerations would facilitate demand-led breeding and make varietal selection in sweetpotato breeding programs more effective. & COPY; 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

BACKGROUNDIt has recently become increasingly evident that banana projects in Uganda need to consider consumer preferences as part of the breeding process to increase the acceptability of new cultivars. A trained panel used quantitative descriptive analysis (QDA) as a tool to assess the sensory characteristics of 32 cooking bananas (matooke). The aim was to investigate which sensory characteristics best describe matooke. RESULTSFourteen descriptors were generated. The preferred attributes of matooke were high-intensity yellow color, homogeneous distribution of yellow color, good matooke aroma, highly moldable by touch, moist and smooth in the mouth. Analysis of variance revealed significant differences in the yellowness, homogeneity of color, firmness, moistness, smoothness, matooke aroma, hardness, and moldability across the genotypes (P < 0.05). Principal component analysis (PCA) showed strong positive correlations between yellowness and homogeneity of the color (R = 0.92). Smoothness in the mouth and moldability by touch were strongly and positively correlated (R = 0.88). Firmness in the mouth was well predicted by hardness to touch (R-2 = 0.85). The matooke samples were ranked into two sensory clusters by agglomerative hierarchical clustering (AHC). CONCLUSIONThe study showed attribute terms that could be used to describe matooke and also revealed that QDA may be used as a tool during the assessment and selection of new cooking banana hybrids to identify relevant sensory attributes because of its ability to discriminate among the banana hybrids. (c) 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

BACKGROUNDThe assessment of user acceptability in relation to crop quality traits should be a full part of breeding selection programs. Our methodology is based on a combination of sensory approaches aiming to evaluate the sensory characteristics and user acceptability of root, tuber and banana (RTB) varieties. RESULTSThe four-stepped approach links sensory characteristics to physicochemical properties and end-user acceptance. It starts with the development of key quality traits using qualitative approaches (surveys and ranking) and it applies a range of sensory tests such as Quantitative Descriptive Analysis with a trained panel, Check-All-That-apply, nine-point hedonic scale and Just-About-Right with consumers. Results obtained on the same samples from the consumer acceptance, sensory testing and physicochemical testing are combined to explore correlations and develop acceptability thresholds. CONCLUSIONA combined qualitative and quantitative approach involving different sensory techniques is necessary to capture sensory acceptance of products from new RTB clones. Some sensory traits can be correlated with physicochemical characteristics and could be evaluated using laboratory instruments (e.g. texture). Other traits (e.g. aroma and mealiness) are more difficult to predict, and the use of a sensory panel is still necessary. For these latter traits, more advanced physicochemical methods that could accelerate the breeding selection through high throughput phenotyping are still to be developed. (c) 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Introduction - Consumers generally underline a gap between the expected quality of the fruit and its perceived one. For mangoes consumed in Europe, mainly imported by sea, this gap can be explained by the decrease of quality during cold storage and the lack of control of the heterogeneity of green mature mango batches. Objectives - To adapt the mode of transport according to the fruit characteristics at the mature green stage, the impact of cold storage in relation to fruit density was assessed on compounds of sensory and nutritional interest. Materials and methods - Low- and high-density fruit were stored for 0 (control) and 15 days at 9 degrees C (cold storage, CS), and then at 20 degrees C until later stages of consumption. Results and discussion - During cold storage, fruit slowly softened and starch was partially hydrolyzed into soluble sugars. After the fruits were transferred to 20 degrees C, the ratio of sucrose to glucose plus fructose increased in CS fruit, whereas it decreased in control fruit during ripening, regardless of the density class. No effect of cold storage was observed on compounds of sensory interest, on the contrary of beta-carotene, which was twice lower in CS fruit. High-density ripe fruit had higher total soluble sugar, beta-carotene, and trans-beta-ionone contents than those of low density, regardless of the storage conditions. Conclusions - The main differences in the sensory quality of mango at the consumption stage were linked to the difference in fruit density at the mature green stage, while the differences in beta-carotene would be associated to cold storage.

Prioritizing sensory attributes and consumer evaluation early in breeding trials to screen for end-user preferred traits could improve adoption rates of released genotypes. In this study, a lexicon and protocol for descriptive sensory analysis (DSA) was established for sweetpotato and used to validate an instrumental texture method for which critical values for consumer preference were set. The study comprised several phases: lexicon development during a 4-day workshop; 3-day intensive panel training; follow-up virtual training, evaluation of 12 advanced genotypes and 101 additional samples from two trials in 2021 by DSA and instrumental texture analysis using TPA double compression; and DSA, instrumental texture analysis and consumer acceptability tests on 7 genotypes in on-farm trials. The established sweetpotato lexicon comprising 27 sensory attributes enabled characterization and differentiation of genotypes by sensory profiles. Significant correlation was found between sensory firmness by hand and mouth with TPA peak positive force (r = 0.695 and r = 0.648, respectively) and positive area (r = 0.748, r = 0.715, respectively). D20, NAROSPOT 1, NASPOT 8, and Umbrella were the most liked genotypes in on-farm trials (overall liking = 7). An average peak positive force of 3700 gf was proposed as a minimum texture value for screening sweetpotato genotypes, since it corresponded with at least 46 % of consumers perceiving sweetpotatoes as just-about-right in firmness and a minimum overall liking of 6 on average. Combining DSA with instrumental texture analysis facilitates efficient screening of genotypes in sweetpotato breeding programs.

Mangifera indica species presents a wide varietal diversity in terms of fruit size and morphology and also of physicochemical and organoleptic properties of the pulp. In Senegal, in addition to the well-known export varieties, such as 'Kent', local varieties have been little studied particularly during ripening. This study aims to propose prediction models integrating variables deduced from varietal characteristics. Five mango varieties ('Diourou', 'Papaye', 'Sierraleone', 'Boukodiekhal' and 'Sewe') endemic to Senegal were characterized at harvest and followed during ripening storage. Caliber parameters were determined at green-mature stage as well as storage (25 °C) weight losses. Considering the 'ripening storage time' (RST) variable as ripeness level index, intra-varietal prediction models were built by multi-linear regression (R2 = 0.98) using pulp pH, soluble solid content (SSC) and Hue angle. In addition to these physicochemical parameters, variety-specific size, shape and weight loss parameters, were additional variables in multi-linear models (R2 = 0.97) for multi-varietal prediction of RST. Results showed that storage time, which was the most influential factor on the pH, SSC and Hue, can be used as a response for varietal prediction of mango ripening. As a decision support tool, theses statistical models, validated on two seasons, will contribute to reduce post-harvest losses and enhance mango value chain through a better ripening process monitoring.

SummaryBoiled pulp is a major form of consumption for plantain. We assessed instrumental (puncture test and texture profile analysis) and sensory texture attributes of 13 plantain cultivars, two cooking hybrids and one dessert banana at different stages of ripeness after cooking in boiling water. Firmness, chewiness, stickiness, mealiness, sweetness and moistness described sensory variability, which was greater between stages of ripeness than between types of cultivars. Firmness and chewiness were well‐predicted by instrumental force and hardness (r2 > 0.72), and by soluble solid and dry matter content (r2 > 0.85). Complementary sensitivity analysis revealed that a pulp puncture force or a hardness of at least 2.1 N or of 0.3 N/mm2 was needed before a difference in firmness or chewiness could be perceived; a Brix of 3.7 was required to ensure a detectable difference in sweetness. Rheological and biochemical predictors can be useful for breeders for high‐throughput phenotyping.

β-carotene content is one of the main factors that determines the nutritional quality and orange color of ripe mango fruit. It is known as the pro vitamin A, and it is generally the predominant carotenoid in ripe mango. Currently, fresh fruit sector demands a reliable nondestructive indicator to better predict the nutritional quality of ripe mango, and especially in terms of β-carotene content. Fruit density has been applied in horticulture sector to predict dry matter or maturation stage of mango. It would interesting to know if it can be a relevant non-destructive indicator of the carotenoid content in fruits Therefore, the aim of this study was to identify the relationship between mango density and β-carotene content at the ripe stage. Mangoes (cv. 'Kent'), from Ivory Coast, Peru, and Brazil were selected from a mango importer in France at a green mature stage (day 0). Then, all mangoes were ripened at 18 °C and 80% of relative humidity. Color of ripe mango pulp and β-carotene contents were assessed 11 and 15 days after (day 0) corresponding to a ripe stage of consumption. Fruit density was significantly correlated to the pulp color and β-carotene contents of ripe mangoes, whatever the origin (p-value < 0.05). Mango fruits with a high density were characterized by high value of Chroma, lower value of Hue angle, and high contents in β-carotene. β-carotene content was found in range of [92-307 μg.100g.FM-1] for low density mango, and in range of [365-924 μg.100g.FM-1] for high density mango. So, fruit density, measured at the green-mature stage, could be a reliable indicator to predict the nutritional quality of mango fruit at the ripening stage, which is extremely useful for fresh fruit sector for grading or sorting mango fruit early in the supply chain.

There is constant demand from the fresh fruit sector for reliable non-destructive indicators to better predict the sensory quality of fruits. The aim of this study was to evaluate the relevance of density to predict mango sensory quality at the ripening stage. Models were built from density and the day of maturation to predict physicochemical indicators of sensory quality. Density of mangoes cv. 'Kent' from Peru, Ivory Coast, and Brazil were assessed at the green mature stage. Sensory characteristics (sweetness, sourness, mango aroma, and firmness intensity) and physicochemical parameters (dry matter content, soluble solids content, total soluble sugar content, titratable acidity, hue angle, and firmness) were assessed on ripe mangoes after storage at 18 degrees C. Mangoes with a higher density had a significantly higher intensity of sweetness, mango aroma, and lower intensity of firmness than mangoes with the lower density, whatever the origin. A minimum difference in density of 0.03-0.04 g mL(-1) was needed to detect significant sensory differences. Physicochemical parameters associated with sensory quality were predicted by density whatever the origin of the fruit. An increase of 0.01 g mL(-1) in density respectively led to an increase of 0.5% in dry matter content, total soluble sugar content, and 0.5 degrees Brix in soluble solid content, and a decrease in hue angle of about 2 degrees to 2.5 degrees. A density threshold of 1.000 g mL(-1) could be applied and easily used to sort heterogeneous batches of mangoes early in the supply chain according to their sensory potential.

Banana green life (GL) is the time between harvesting and the start of natural ripening. GL could be considered as a major quality criterion, as it defines whether or not a fruit is suitable for export and marketing. The ending of GL, when climacteric crisis occurs, is characterized by autocatalytic ethylene production. Ethylene synthesis and regulation are described. The main methods for determining GI are based on detecting the CO2 peak or a decrease in green pigments, using a spectrometer (NDVI) for the latter. Temperatures during fruit growth and Black Sigatoka disease are the main pre-harvest factors affecting GL. The former can be managed by applying the thermal sums concept and the latter by adequate field practices. The effects of the main exogenous storage parameters (storage temperature, relative humidity, concentration of ethylene and O-2/CO2 ratio in the atmosphere) could be modelled in some cases. The most effective solutions for extending GL rely on either developing coatings using new preservative compounds, or designing packaging capable of controlling the temperature, CO2/O-2 ratio and ethylene concentrations in the environment close to the fruits. It was demonstrated that 1-MCP is not relevant for increasing GL. A global and integrated approach involving the overall optimization of pre- and post-harvest factors needs to be applied to maintain green fruit until voluntary/artificial ripening is induced.

Organic food products are subjected to high risk of fraud in Europe. Food traceability and authenticity of foods are mainly ensured by administrative means and there is a lack of analytical tools to authenticate organic foods. Hence, we wanted to propose a methodology that could help control and certification bodies for the authentication of organic foods. In this study we demonstrate the robustness of our original approach by comparing bacterial flora of bananas from different farming types, varieties, harvest years and geographical origins. Interestingly, the farming type could be linked to variations in bacterial diversity of bananas even if the geographical origin of bananas had the highest impact. Also, some bacterial groups have demonstrated a higher discriminative power. This opens the perspectives for the development of a food authentication tool based on the use of the natural microbial ecosystem of foods.

The management of banana postharvest conditions is a pre-requisite for adapting the storage, shipping and gassing conditions to the specific physiology of individual hybrid bananas. The cultivar 'CIRAD 925', developed by CIRAD, is a good case study for adapting banana food chain conditions to accommodate the specificities of this hybrid. Indeed, the storage and ripening conditions should be adapted to this cultivar which is tolerant to the black leaf streak disease. The management of temperature and relative humidity during the maturation step is therefore a key challenge to avoid peel splitting and browning. These physiological disorders are exposed here, and associated biochemical factors are described and discussed. The cuticular wax behaviour during ripening at different relative humidities and its impact on peel splitting and browning is associated with the intensity of these disorders.

This report presents results of RTB-ENDURE sub-output 1.3; 'Determining appropriate harvest time for the cooking bananas with intrinsic long shelf-life using physical, chemical and sensory attributes' of the cooking banana business case's Output 1, entitled “Increased access of farmers to cooking banana varieties with preferred quality attributes and intrinsic long shelf life traits”. The worked aimed at reducing postharvest losses for cooking banana while modulating harvest stage for green life extension. The originality of this investigation was to evaluate the putative impact of fruit stage of harvest onto its potential storage life and eating quality. The optimal harvest stage was evaluated by coupling three antagonist parameters, namely fruit diameter, green life, and eating quality, to optimize harvest stage of the variety Kibuzi in specific edapho-climatic conditions of Rakai and Isingiro districts in southwestern Uganda. A temperature record was considered in both sites between flowering and harvest. The interval between flowering and harvest (IFH) of Kibuzi banana variety was used as a quantitative explanatory variable, and the site location (Rakai at 1270 masl vs Isingiro at 1440 masl) was used as a qualitative one. Since the sites were at different altitudes, two Tynitag temperature data loggers were installed to record temperatures. Fruits size, dry matter, fruit firmness, total soluble solids, titratable acidity and sensory attributes were recorded at four harvest stages: 112, 126, 138, 152 days and 111, 125, 137, 151 days after flowering. The evolution of three parameters; diameter of fruit, green life and overall acceptability of the end-product - Matooke - were simulated for 110 to 155 days range, leading to the identification of a range of optimal harvest ages for variety Kibuzi in Rakai at between 133 to 142 days and 133 to 150 days for Isingiro. The prediction of the optimal harvest stage will remain only valid for the two locations without taking into account thermal sum for establishing a strong relationship between fruit age in degree.days and green life. Given the respective altitudes at Rakai and Isingiro, it implies that the two edapho-climatic conditions were not so different in terms of on field temperature. With some more diverse thermal conditions in the experimental sites (lowland vs highland with at least 3°C needed between sites), the thermal sum concept will be even more precise for the prediction of the optimal harvest stage for bananas, regardless the location site (lowland, highland, with hot or cool local conditions). Such original multi-criteria approach (agro-morphological, physiological traits, and end-product sensory attributes) was relevant for the prediction of the optimal harvest stage, in order to reduce banana postharvest losses during transport and until Matooke preparation by end-users. Such innovative methodology can be applied to some other banana culinary recipes and end-uses. (Resume d'auteur)

La banane dessert, qui est la premiere ressource agricole aux Antilles, doit faire face d'une part a une forte concurrence avec les pays d'Afrique et d'Amerique latine et d'autre part a une pression parasitaire importante, notamment due aux maladies foliaires des bananiers. Dans ce contexte, la creation de varietes de banane dessert presentant une resistance aux principales maladies et ravageurs du bananier et de nouvelles caracteristiques de qualite sensorielle apparait comme une voie privilegiee pour accroitre la viabilite et la durabilite de la filiere. Dans le cadre du programme d'amelioration varietale de la banane dessert cree par le CIRAD et base aux Antilles, mon role a ete de contribuer a son efficience en participant a la production (1) de methodes de phenotypage de la qualite sensorielle et des desordres physiologiques, (2) d'outils d'aide a la selection varietale et (3) de connaissances sur l'elaboration de la qualite sensorielle du fruit. Pour cela, mon projet scientifique s'est articule autour de deux axes de recherche. Le premier axe visait l'identification d'indicateurs pertinents de la qualite sensorielle de la banane dessert tres peu abordee auparavant. Ces indicateurs ainsi que leur seuil d'acceptabilite ont ete evalues a partir d'une gamme diversifiee de genotypes. Il a pu etre demontre notamment l'importance des parametres d'acidite dans l'expression de plusieurs descripteurs sensoriels (acide, sucre, certains aromes, fermete). Le second axe de recherche visait a analyser l'impact du genotype et des pratiques culturales sur l'evolution des composes d'interet sensoriel et sur l'expression de desordres physiologiques de la banane. En s'appuyant a la fois sur la conduite d'experimentations au champ et la construction de modeles statistiques ou mecanistes, des elements ont ete apportes sur les origines de la variabilite mesuree qu'elle soit genotypique ou liee aux conditions environnementales. Des outils sont maintenant disponibles pour predire le potentiel de conservation, l'apparition de la frisure et l'acidite chez differents genotypes selon les conditions de culture. Concernant l'acidite, les modeles developpes ont permis d'identifier les processus physiologiques cles qui pilotent ce critere. L'amelioration de la qualite sensorielle de la banane dessert est simulee a partir de scenarii mettant en avant une optimisation de la selection varietale et des pratiques culturales. Mes perspectives scientifiques sont de poursuivre les travaux sur la comprehension des processus pilotant la qualite sensorielle des fruits mais cette fois-ci soumis a des itineraires post-recolte (impliques dans la conservation, la maturation, ou la transformation des fruits). Les travaux contribueront a (1) limiter les pertes et gaspillages des fruits par une optimisation de ces itineraires, et (2) renforcer la securite alimentaire des pays du Sud en participant a l'identification de criteres de selection specifiques aux usages alimentaires de ces pays.

Peel splitting is a major physiological disorder affecting post-harvest banana quality. This phenomenon occurs only 3–6 days after ripening induction in specific cultivars such as cv. 925 when stored in saturating humidity conditions. In these conditions, Cavendish cultivars (Grande Naine, cv. GN) are not susceptible to splitting. Cvs. 925 and GN were thus investigated to detect possible determinants associated with splitting. Splitting intensity was tentatively found to be associated with an inverse water flux at high relative humidity (RH) through an osmotic peel to pulp water flux resulting from the higher sugar content in the pulp than in the peel. Rheological properties were measured, and although the peel resistance and elasticity in cv. 925 were surprisingly higher than in cv. GN, saturating humidity conditions (100 % RH) substantially reduced the peel resistance. However, the peel epicuticular wax in cv. 925 was clearly thinner than that in cv. GN, thus leading to limitation of peel hydration in cv. GN. Peel splitting in cv. 925 was also associated with a boost in respiration, an increase in oxidative stress markers (H2O2), resulting in an increase in cellular damage markers (MDA, PEL). Overall, our results suggest that peel splitting at high RH in cv. 925 is related to fast decrease peel water content and the induction of high oxidative stress damage.