Search

Near-infrared spectroscopy (NIRS) is a key non-destructive technique for rapid assessment of the chemical properties of food materials. However, a major challenge with NIRS is the mixed physicochemical phenomena captured by the interaction of the light with the matter. The interaction often results in both absorption and scattering of the light. The overall NIRS signal therefore contains information related to the two phenomena mixed. To predict chemical properties such as dry matter, Brix and lipids, light refelction/absorption is used. Therefore, when the aim of the data analysis is to predict chemical components, it is necessary to remove as much as possible the scattering effects from the spectra. Several pre-processing techniques are available to do this, but it is often difficult to decide which one to choose. In this article we present the use of a recently developed pre-processing approach, sequential pre-processing through orthogonalization (SPORT), to improve the predictive power of multivariate models based on NIR spectra of food materials. The SPORT approach utilizes sequential orthogonalized partial least square regression (SOPLS) for the fusion of data blocks corresponding to several spectral preprocessing techniques. The results were compared with commonly used pre-processing techniques in the analysis of food materials by NIRS. The comparison was made by analyzing 5 different datasets comprised of apples, apricots, olive oils and grapes associated with chemical properties such as dry matter (DM), Brix, lipids and citric acid. The datasets were from both reflection and transmission measurements. The results showed that the fusion-based pre-processing methodology is an ideal choice for pre-processing of NIRS data. For four out of five datasets, the prediction accuracies (high R2pred and low RMSEP) were improved. The improvement led to as much as a 20 % increase in R2pred and a 25 % decrease in RMSEP compared to the standard 2nd derivative pre-processing. The pre-processing fusion was more effective for the reflection mode compared to the transmission mode. Multiple pre-processing techniques provided complementary information, and therefore, their fusion using the SPORT approach improved the model performance. The methodology is not only applicable to food materials but can in fact be used as a general pre-processing approach for all types of modeling of spectral data.

Assessing the impact of variation in chloroplast and mitochondrial DNA (collectively termed the plasmotype) on plant phenotypes is challenging due to the difficulty in separating their effect from nuclear derived variation (the nucleotype). Haploid inducer lines can be used as efficient plasmotype donors to generate new plasmotype-nucleotype combinations (cybrids)(Ravi et al., 2014). We generated a panel comprising all possible cybrids of seven Arabidopsis thaliana accessions and extensively phenotyped these lines for 1859 phenotypes under stable and fluctuating conditions. We show that natural variation in the plasmotype results in additive as well as epistatic effects across all phenotypic categories. Plasmotypes which induce more additive phenotypic changes also cause more significant epistatic effects, suggesting a possible common basis for both additive and epistatic effects. On average epistatic interactions explained twice as much of the variance in phenotypes as additive plasmotype effects. The impact of plasmotypic variation was also more pronounced under fluctuating and stressful environmental conditions. Thus, the phenotypic impact of variation in plasmotypes is the outcome of multilevel Nucleotype X Plasmotype X Environment interactions and, as such, the plasmotype is likely to serve as a reservoir of variation which is only exposed under certain conditions. The production of cybrids using haploid inducers is a quick and precise method for assessing the phenotypic effects of natural variation in organellar genomes. It will facilitate efficient screening of unique nucleotype-plasmotype combinations to both improve our understanding of natural variation in nucleotype plasmotype interactions and identify favourable combinations to improve plant performance.

Comment les plantes poussent-elles ? Le nombre d’or est-il omniprésent dans leur architecture ? Leurs structures sont-elles fractales ? Ont-elles un langage ? Une grammaire ? Il y a du vrai dans ces questions, mais il faut aller plus loin. Il faut identifier, modéliser et simuler le rôle des organes d’une plante dans sa croissance et leur fonctionnement.La modélisation des plantes fait l’objet de recherches à l’interface de disciplines biologiques (botanique, agronomie, génétique, écophysiologie) des mathématiques appliquées et de l’informatique. Elle permet de créer les modèles de développement de l’architecture des plantes, mais aussi les modèles de production végétale, expression de la croissance des organes.Les auteurs de cet ouvrage nous exposent les fondements biologiques, mathématiques et informatiques qui permettent d’exprimer le fonctionnement des bourgeons, la production photosynthétique de la biomasse et sa répartition dans les organes d’une plante. La simulation de la croissance des plantes devient alors possible sous la forme de modèles dynamiques et sa représentation, sous la forme d’images de synthèse. Ces modèles autorisent de multiples applications. De nombreuses plantes sont présentées (herbacées, arbustes et arbres) ; leur modélisation a une visée pratique en agriculture, en gestion des ressources naturelles et de l’environnement et en représentation des paysages.Rédigé par des botanistes, des agronomes, des mathématiciens et des informaticiens, cet ouvrage collectif est le fruit de quarante années de recherches conduites par Philippe de Reffye et ses collègues, avec des collaborations scientifiques en France, en Hollande, en Chine et en Afrique. Il s’adresse aux chercheurs, enseignants et étudiants en biologie, en agronomie et en sciences de la vie, aux architectes paysagistes, écologues, qui s’intéressent à la croissance des plantes, à leur modélisation et à leur représentation. Il s’adresse aussi à toutes les communautés des sciences exactes (mathématiciens, informaticiens, physiciens) intéressées par la simulation du vivant. Ce livre paraîtra au format papier et PDF en juin 2017.

This paper proposes a novel knowledge-and-data-driven modeling (KDDM) approach for simulating plant growth that consists of two submodels. One submodel is derived from all available domain knowledge, including all known relationships from physically based or mechanistic models; the other is constructed solely from data without using any domain knowledge. In this work, a GreenLab model was adopted as the knowledge-driven (KD) submodel and the radial basis function network (RBFN) as the data-driven (DD) submodel. A tomato crop was taken as a case study on plant growth modeling. Tomato growth data sets from twelve greenhouse experiments over five years were used to calibrate and test the model. In comparison with the existing knowledge-driven model (KDM, BIC=1215.67) and data-driven model (DDM, BIC=1150.86), the proposed KDDM approach (BIC=1144.36) presented several benefits in predicting tomato yields. In particular, the KDDM approach is able to provide strong predictions of yields from different types of organs, including leaves, stems, and fruits, even when observational data on the organs are unavailable. The case study confirms that the KDDM approach inherits advantages from both the KDM and DDM approaches. Two cases of superposition and composition coupling operators in the KDDM approach are also discussed. (C) 2015 Elsevier B.V. All rights reserved.