Bioactive peptides have significant potential for applications in pharmaceuticals, the food industry, and cosmetics due to their wide spectrum of biological activities. However, their pronounced structural and functional heterogeneity complicates the classification and prediction of biological activity. This study uses data from the PepLab platform, comprising 2748 experimentally confirmed bioactive peptides distributed across 15 functional classes, including ACE inhibitors, antimicrobial, anticancer, antioxidant, toxins, and others. For each peptide, the amino acid sequence and key physicochemical descriptors are provided, calculated via the integrated DMPep module, such as GRAVY index, aliphatic index, isoelectric point, molecular weight, Boman index, and sequence length. The dataset exhibits class imbalance, with class sizes ranging from 14 to 524 peptides. An innovative methodology is proposed, combining descriptive statistical analysis, structural modeling via DEMATEL, and structural equation modeling with neural networks (SEM-NN), where SEM-NN is used to capture complex nonlinear causal relationships between descriptors and functional classes. The results of these dependencies are integrated into a multi-class machine learning model to improve interpretability and predictive performance. Targeted data augmentation was applied to mitigate class imbalance. The developed classifier achieved predictive accuracy of up to 66%, a relatively high value given the complexity of the problem and the limited dataset size. These results confirm that integrating structured dependency modeling with artificial intelligence is an effective approach for functional peptide classification and supports the rational design of novel bioactive molecules.

An experimental study is presented on the possibility of using the fluorescence from organic dyes as a broadband light source together with a monochromator for applications in excitation–emission matrix (EEM) fluorescence spectroscopy. A high-power single-chip light-emitting diode (LED) was chosen as an excitation source with a central output wavelength at 365 nm to excite a fluorescent solution of Coumarin 1 dye dissolved in ethanol. Two excitation configurations were investigated: direct excitation from the LED and excitation through an optical-fiber-coupled LED. A Czerny–Turner monochromator with a diffraction grating was used for the spectral tuning of the fluorescence. A simple method was investigated for increasing the efficiency of the excitation as well as the fluorescence signal collection by using a diffuse reflector composed of barium sulfate (BaSO₄) and polyvinyl alcohol (PVA). As research objects, extra-virgin olive oil (EVOO), Coumarin 6 dye, and Perylene, a polycyclic aromatic hydrocarbon (PAH), were used. The results showed that the light-emitting-diode-induced fluorescence was sufficient to cover the losses on the optical path to the monochromator output, where a detectable signal could be obtained. The obtained results reveal the practical possibility of applying the fluorescence from dyes as a light source for food system analysis by EEM fluorescence spectroscopy.

A significant quantity of agro-industrial waste is generated globally across various agricultural sectors and food industries. Composed primarily of cellulose, hemicellulose, and lignin—known as lignocellulosic materials—this waste holds significant potential and can be repurposed as a nutrient-rich substrate for mushroom cultivation. Therefore, mushroom cultivation can be regarded as a promising biotechnological approach for the reduction and valorization of agro-industrial waste. This investigation is the first to explore the utilization of agro-industrial waste- and by-products for the cultivation of Epicoccum nigrum for the production of extracts with valuable biological activities. The logistic curve and autocatalytic growth models were applied to study the kinetics of the growth process on wheat bran, sunflower cake, wheat straw, pine sawdust, and steam-distilled lavender straw substrates. Through mathematical modeling, the optimal composition of a nutrient medium containing the selected substrates was determined and successfully validated in experimental conditions. Biologically active water extracts were obtained after solid-state cultivation with α-amylase and cellulase activity up to 10.6 ± 0.6 U/mL and 0.52 ± 0.03 U/g, respectively. The extracts exhibited antimicrobial activity against fungal strains from six different species, and the most susceptible was the phytopathogen Sclerotinia sclerotiorum, with a minimum inhibitory concentration of 0.156–0.313 mg/mL.