Chemometric method development for metabolomic analyses for food security and authentication

该项目旨在开发新的方法来分析代谢组学和转录组学的大型数据集，以进一步了解这将有助于全球粮食安全，减少大豆进口，并有利于欧盟经济和环境。考虑到气候变化，了解一些植物抵御干旱和疾病的机制至关重要。将对时间序列数据进行分析，以了解这些因素如何影响豆科植物（如豌豆）的生长。其目的是使用复杂的统计和模式识别技术分析来自遭受干旱和镰刀菌感染的植物的数据。特别地，将从植物DNA获得的联合收割机数据与来自代谢组学的数据结合，以鉴定使植物对干旱和/或疾病具有抗性的基因，以及来自实验的数据，以鉴定植物内的化学物质，所述化学物质提供关于参与对真菌感染和干旱的抗性的代谢途径的信息。基于质谱和核磁共振（NMR）的技术将用于非靶向代谢组学分析，并结合QTL（数量性状基因座）作图实验的数据。这些组学数据集的融合并不是微不足道的，基于相关或级联的新的多变量方法以及基于路径的方法将被开发。将不同技术的数据结合起来将需要开发新技术，其结果可能导致快速识别更具适应性的作物品种。该项目的另一个目的是提高对黑水虻（BSF）幼虫的营养需求的理解，以便将食物废物流最佳地转化为昆虫生物质。昆虫，特别是苍蝇，有可能为动物饲料提供蛋白质来源，作为日益昂贵的进口大豆和鱼粉的替代品，从而改善碳足迹并减少过度捕捞。蝇幼虫是鱼、鸡和猪饮食的天然成分，可以在一系列有机废物上生长，减少多达60%的废物量，为废物管理和环境提供额外的好处。为了实现这一目标，通过微生物降解（厌氧消化）从可用的食物垃圾中获得的底物将使用质谱法和NMR进行分析，并与幼虫蛋白质产量相关，以确定氮、蛋白质和脂肪等营养素的最低所需含量。还将对各种生物降解食品废物流的营养质量进行分析，并对各种废物流的组合进行建模，以优化饲料基质的营养质量。与利用商品价格来确定满足法规要求的最经济的动物饲料组成一样，将开发原型软件，以整合现有废物流中基本营养素含量的信息，从而优化基质生产，从而优化昆虫蛋白质产量。关键词化学计量学，统计学，数据分析，模式识别，数学建模

This project aims to develop novel methods for the analysis of large datasets from metabolomics and transcriptomics in order to further understanding that will contribute to worldwide food security, reduce soya importation and benefit the EU economy and the environment. In view of climate change, understanding the mechanisms that allow some plants to withstand drought and disease is vital. Time series data will be analysed to see how these factors affect the growth of leguminous plants, such as peas. The aim is to analyse data from plants subjected to both drought and Fusarium-infection using sophisticated statistical and pattern-recognition techniques. In particular, to combine data obtained from the plant DNA with data from metabolomics to identify the genes making a plant resistant to drought and/or disease with data from experiments to identify the chemicals within the plant that provide information on the metabolic pathways involved in resistance to fungal infection and drought. Mass-spectrometry and nuclear magnetic resonance (NMR)-based techniques will be used in non-targeted metabolomic analysis combined with data from QTL (Quantitative Trait Locus) mapping experiments. The fusion of these 'omics datasets is not trivial novel multivariate methods based on correlation or concatenation as well as pathway-based methods will be developed. The combination of data from different technologies will require the development of new techniques and the results could lead to the rapid identification of more resilient crop varieties. Another aim of this project is to improve understanding of the nutritional requirements of black soldier fly (BSF) larvae for optimal conversion of food waste streams to insect biomass. Insects, especially flies, have the potential to provide a protein source for animal feed as an alternative to increasingly expensive imported soya and fishmeal, thereby improving the carbon footprint and reducing overfishing. Fly larvae are a natural component of the diet of fish, chicken and pigs and can be grown on a range of organic wastes, reducing the volume of that waste by up to 60%, providing an additional benefit to waste management and the environment. To achieve this, the substrates obtained from available food waste products via degradation by microorganisms (anaerobic digestion), will be profiled using mass spectrometry and NMR and correlated with larval protein yield to allow the minimum required content of nutrients such as nitrogen, protein and fats to be determined. Analysis of the nutritional quality of various bio-degraded food waste streams will also be performed and methods to model the combination of various waste streams to optimize the nutritional quality of feed substrate developed. In the same way that commodity prices are used to determine the most economical composition of animal feed that satisfies regulations, prototype software will be developed to integrate information on the content of essential nutrients from available waste streams to optimise substrate production and hence insect protein yield.KeywordsChemometrics, statistics, data analysis, pattern recognition, mathematical modelling