Sorting hidden patterns in nanoparticle performance for glioblastoma using machine learning algorithms

Cationic compounds have been described to readily penetrate cell membranes. Assigning positive charge nanosystems, e.g. lipid nanoparticles, has identified as a key feature promote electrostatic binding and design ligand-based constructs for tumour targeting. However, their intrinsic high cytotoxicity hampered biomedical application. This paper seeks establish which cationic properties are compelling interface modulation, in order improve the of targeted nanoparticles against glioblastoma. How can features (e.g. nature, structure, conformation) shape efficacy outcomes? In quest safer alternative compounds, we evaluate effects two novel glycerol-based lipids, GLY1 GLY2, on architecture performance nanostructured carriers (NLCs). These molecules, composed alkylated chains glycerol backbone, differ only polar head proved be efficient reversing zeta potential nanosystems values. The use unsupervised supervised machine learning (ML) techniques unraveled structural similarities: spite common exhibited better increasing cytotoxicity, while decreasing particle size. Furthermore, NLCs containing showed favorable hemocompatible profile, well an improved uptake by cells. Summing-up, circumvents surfactant, CTAB, is effective at glioblastoma uptake, exhibits encouraging anticancer activity. Moreover, ML strongly incited formulation optimization.