Integrating infrared spectroscopy, metabolomics, and network pharmacology to investigate differential metabolites of volatile compounds in porcini mushrooms

Wild mushrooms are a natural diet. They are a key source of nutritious food and healthful compounds. This study used infrared (IR) spectroscopy and GC-MS to compare five porcini species. It aimed to trace the species information. Network pharmacology is used to analyze targets and pathways of porcini mushroom compounds. They intervene in the gut microbiota to modulate colorectal cancer, obesity, and diabetes. The results showed that IR spectra visualized the chemical compositional differences among porcini species. Two-dimensional correlation spectroscopy (2DCOS) combined with residual convolutional neural network (ResNet) modeling successfully identified and validated the differences and established a model for the rapid identification of porcini species (accuracy=1.0, loss function close to 0). GC-MS analyses revealed significant differences in the volatile components (VOCs) of the five porcini species, and 15 bioactive components (phenols and terpenes) were screened as potential markers. Network pharmacology suggests that potential chemical markers affect the gut microbiota. They may modulate colorectal cancer, obesity, and diabetes through "multi-targets-multi-pathways." This study lays the foundation for the quality evaluation study of porcini mushrooms and provides a basis for the development of healthcare products.