No abstract

:  Although tomato flavor has not been a major goal for breeders, nowadays it becomes important as it is a subject of consumer complaint. A better knowledge of tomato consumer preferences, at the European level, should provide the basis for improvement of fruit quality and for market segmentation. In the framework of a large European project, 806 consumers from 3 countries, The Netherlands, France, and Italy, were presented with a set of 16 varieties representing the diversity of fresh tomato offer in order to evaluate their preferences. In parallel, sensory profiles were constructed by expert panels in each country. Preference maps were then constructed in each country revealing the structure of consumer preferences and allowing identification of the most important characteristics. Then a global analysis revealed that preferences were quite homogeneous across countries. This study identified the overall flavor and firmness as the most important traits for improving tomato fruit quality. It showed that consumer preferences from different European countries, with different cultures and food practices, are segmented following similar patterns when projected onto a common referential plan. Moreover, the results clearly showed that diversification of taste and texture is required to satisfy all consumers’ expectations as some consumers preferred firm tomatoes, while others preferred melting ones and were more or less demanding in terms of sweetness and flavor intensity. Detailed comparisons also showed the importance of the fruit appearance in consumer preference. Practical Application:  The consumer preferences for fresh market tomato were studied in 3 European countries. The main descriptors for further breeding for consumer satisfaction were identified. Four clusters of consumers were identified in the overall analysis, the 3 countries contributing the same way to each cluster. The impact of appearance in the preferences was also underlined.

Object detection from aerial and satellite remote sensing images has been an active research topic over the past decade. Thanks to the increase in computational resources and data availability, deep learning-based object detection methods have achieved numerous successes in computer vision, and more recently in remote sensing. However, the ability of current detectors to deal with (very) small objects still remains limited. In particular, the fast detection of small objects from a large observed scene is still an open question. In this work, we address this challenge and introduce an enhanced one-stage deep learning-based detection model, called You Only Look Once (YOLO)-fine, which is based on the structure of YOLOv3. Our detector is designed to be capable of detecting small objects with high accuracy and high speed, allowing further real-time applications within operational contexts. We also investigate its robustness to the appearance of new backgrounds in the validation set, thus tackling the issue of domain adaptation that is critical in remote sensing. Experimental studies that were conducted on both aerial and satellite benchmark datasets show some significant improvement of YOLO-fine as compared to other state-of-the art object detectors.

The R package FAMT (factor analysis for multiple testing) provides a powerful method for large-scale significance testing under dependence. It is especially designed to select differentially expressed genes in microarray data when the correlation structure among gene expressions is strong. Indeed, this method reduces the negative impact of dependence on the multiple testing procedures by modeling the common information shared by all the variables using a factor analysis structure. New test statistics for general linear contrasts are deduced, taking advantage of the common factor structure to reduce correlation and consequently the variance of error rates. Thus, the FAMT method shows improvements with respect to most of the usual methods regarding the non discovery rate and the control of the false discovery rate (FDR). The steps of this procedure, each of them corresponding to R functions, are illustrated in this paper by two microarray data analyses. We first present how to import the gene expression data, the covariates and gene annotations. The second step includes the choice of the optimal number of factors, the factor model fitting, and provides a list of selected genes according to a preset FDR control level. Finally, diagnostic plots are provided to help the user interpret the factors using available external information on either genes or arrays.

Handling dependence or not in feature selection is still an open question in supervised classification issues where the number of covariates exceeds the number of observations. Some recent papers surprisingly show the superiority of naive Bayes approaches based on an obviously erroneous assumption of independence, whereas others recommend to infer on the dependence structure in order to decorrelate the selection statistics. In the classical linear discriminant analysis (LDA) framework, the present paper first highlights the impact of dependence in terms of instability of feature selection. A second objective is to revisit the above issue using a flexible factor modeling for the covariance. This framework introduces latent components of dependence, conditionally on which a new Bayes consistency is defined. A procedure is then proposed for the joint estimation of the expectation and variance parameters of the model. The present method is compared to recent regularized diagonal discriminant analysis approaches, assuming independence among features, and regularized LDA procedures, both in terms of classification performance and stability of feature selection. The proposed method is implemented in the R package FADA, freely available from the R repository CRAN.

No abstract

Asymptotic tests are commonly used for comparing two binomial proportions when the sample size is sufficiently large. However, there is no consensus on the most powerful test. In this paper, we clarify this issue by comparing the power functions of three popular asymptotic tests: the Pearson's χ test, the likelihood-ratio test and the odds-ratio based test. Considering Taylor decompositions under local alternatives, the comparisons lead to recommendations on which test to use in view of both the experimental design and the nature of the investigated signal. We show that when the design is balanced between the two binomials, the three tests are equivalent in terms of power. However, when the design is unbalanced, differences in power can be substantial and the choice of the most powerful test also depends on the value of the parameters of the two compared binomials. We further investigated situations where the two binomials are not compared directly but through tag binomials. In these cases of indirect association, we show that the differences in power between the three tests are enhanced with decreasing values of the parameters of the tag binomials. Our results are illustrated in the context of genetic epidemiology where the analysis of genome-wide association studies provides insights regarding the low power for detecting rare variants.

No abstract