Data analysis for Big Vibrational Spectroscopic Data

Due to an increasing focus on heathy diets and low-fat products, huge amounts of animal fat by-products are processed annually by fat processors and renderers. Most of the produced fat products are used for energy consumption. In order to add value to these low-value products, the LipoFungi project aims at upgrading Norwegian animal slaughter fat by-products to high-value fatty acids by fungal fermentation processes. In a pilot study, we have shown that oleaginous fungi from the phylum Zygomycetes can convert saturated fatty acids from animal fat by-products containing triacylglycerols (SAT-TAGs), into polyunsaturated fatty acids containing triacylglycerols (PUFA-TAGs). The lions share of the fat by-products that are produced in Norway and Europe today are either edible fats fitting for human consumption or so-called Category 3 fat that is suitable for animal feed. Therefore, depending on the type of fat by-product, high-value fatty acids produced from Category 3 fat rest materials or edible fat rest materials could be used for feed or for human consumption, respectively. The project addresses the most central biological and technological research challenges for developing a sustainable process for lipid upgrading from animal rest fat. It performs research for developing fermentation technology for submerged and solid-state fermentation, process optimization and upscaling. The project evaluates the final product and performs a preliminary economic evaluation of the process. The project makes use of the NordBioLab infrastructure facilities at NMBU and constitutes a multi-disciplinary and cross-institutional collaboration involving all competencies required for such a complex goal.

Vibrational spectroscopy and imaging are applied in live sciences and biomedical sciences for the investigation of biological samples in their native form. The techniques are used for cancer diagnosis, microbial identification and in general for characterization of biological samples. The field is evolving quickly and the scientific community of biomedical vibrational spectroscopy is big. It is for example expected that within short time histopathological analysis for cancer diagnosis will be computer-assisted or replaced by vibrational spectroscopic imaging combined with big data analysis. Huge databases of spectroscopic imaging data are currently collected at different centres. The data analytical challenges are the scatter correction of the data and the establishing of powerful classifiers. The Norwegian group at NMBU is renowned for being at the forefront of data analysis of biomedical vibrational spectroscopic data. The French group at Soleil is at the forefront of ground-breaking new instrumental settings for vibrational spectroscopic imaging of biological systems and represents one of the main platforms for vibrational spectroscopic imaging in Europe. The BigSpecData project aims at establishing a long-term collaboration between the Norwegian University of Life Sciences and the SMIS Infrared Spectromicroscopy Beamline at Soleil on Big Data obtained in vibrational spectroscopic Imaging. The collaboration will address data analysis challenges that have been arising during recent years in the field of vibrational spectroscopic imaging of biological materials.