Bayesian Information Fusion for Robust Classification and Quantification of ill-defined Biosamples with Applications to Studies of Microalgae

田纳西大学诺克斯维尔的Frank沃格特教授得到了化学系化学测量和成像（CMI）项目的支持，以开发用于复杂化学和生物样品定性和定量分析的化学计量学方法。具体目标是开发“贝叶斯信息融合”，这是一种新的化学计量学概念，它将来自互补来源的不同类型的信息合并用于样品分类。为了进一步提高分类和量化的准确性，将开发和采用一种新的“合理性分析”软件。作为原理验证，这些新的化学计量学方法将应用于从微藻样品中获得的FTIR光谱数据，目标是将它们用作环境生物探针。复杂化学和生物样品的分析极具挑战性，因为存在大量的分析物及其广泛的浓度范围，和化学计量学数据分析正日益成为一种宝贵的分析化学工具，用于改进此类样品的分类和定量。该项目的结果将通过网络化和交互式软件包“虚拟化学计量学实验室。"

Professor Frank Vogt of the University of Tennessee Knoxville is supported by the Chemical Measurement and Imaging (CMI) Program in the Division of Chemistry to develop chemometric methods for the qualitative and quantitative analysis of complex chemical and biological samples. The specific objective is to develop "Bayesian Information Fusion," a novel chemometric concept which merges different types of information from complementary sources for sample classification. To further enhance the accuracy of the classification and the quantification, a novel "Plausibility Analysis" software will be developed and incorporated. As a proof-of-principle, these novel chemometric methodologies will be applied to FTIR spectroscopic data obtained from microalgae samples with the goal to utilize them as environmental bioprobes.The analysis of complex chemical and biological samples is extremely challenging because of the large number of analytes present and their wide range of concentrations, and chemometric data analysis is increasingly becoming an invaluable analytical chemistry tool for improved classification and quantification of such samples.The project results will be integrated in classroom material by means of the cyber-enabled and interactive software package "Virtual Chemometrics Laboratory."