Machine learning-assisted data-driven optimization and understanding of the multiple stage process for extraction of polysaccharides and secondary metabolites from natural products

Abstract

A machine learning strategy mainly consist of radial basis function neural network and genetic algorithm for predicting and understanding multi-objective extraction process.

Valorization of biomass for food protein via deep eutectic solvent extraction: Understanding the extraction mechanism and impact on protein structure and properties

“…Although there are numerous HBA:HBD combinations that can solubilize proteins, it is currently unclear how all of those properties interact to affect protein extraction. In order to maximize the extraction of biomolecules from plant sources, new strategies incorporating computer-aided technology, such as machine learning, are very helpful when analyzing complicated data with various parameters (Ma et al, 2023;Shekhar et al, 2023). Shi et al (2022) successfully employed a machine learning algorithm to predict DES viscosity by considering the basic properties (i.e., molar mass of HBA and HBD, HBA:HBD molar ratio, and temperature), molecular fingerprint of the HBA:HBD composition, and water content as the model input.…”

“…Although there are numerous HBA:HBD combinations that can solubilize proteins, it is currently unclear how all of those properties interact to affect protein extraction. In order to maximize the extraction of biomolecules from plant sources, new strategies incorporating computer‐aided technology, such as machine learning, are very helpful when analyzing complicated data with various parameters (Ma et al., 2023; Shekhar et al., 2023). Shi et al.…”