In this project we intend to advance in the development of new chemometric methods, especially of multivariate resolution, for massive multi - and megavariate data processing. The development and application of multivariate resolution methods, of which our research groups are main drivers and promoters around the world, is providing workable solutions to many analytical problems stemming from the massive multi- and megavariate characteristics of the acquired data. The reason for this success is that this type of methods allows managing massive and diverse complex information and provides simple representations of the acquired information, using a small set of basic, easy to visualize and interpret contributions with chemical sense. This quality compressed and representative information of the original data can be reused easily as a starting point for further analysis. Another advantage of the proposed methods is the simple structure of its algorithm, which makes it adaptable to very different contexts, and makes this project markedly multidisciplinary. Metabonomic and genomic problems will be resolved specifically, where the information obtained is crucial for the interpretation of the effects of the contaminants and of other environmental stressors on the biological organisms studied. The proposed methods will be used in environmental chemistry problems, in which the mofre important patterns and sources of pollution will be described, together with their composition and geographical distribution. They will be of great importance in the analysis of hyperspectral images, in which the the key point is the visualization of the spatial and spectral information of the components of the samples.
And, finally, the modeling of processes will be addressed in bioanalytical systems of great complexity, for which often there is no established physico-chemical knowledge of the behaviour, and in the PAT environment, where there are still pending challenges related to the interpretability of the phenomena and of the process evolution trajectories, which need more transparent and interpretable control systems preserving the nature of the original measurements and enabling the correct data fusion of diverse origin according to the nature of the process and of the parameters to control.