(AGENPARL) – LONDON (UNITED KINGDOM), mer 16 dicembre 2020
Nanoparticles are useful antimicrobial drugs release systems however some nanoparticles has anti-bacterial activity per se. Their discovery is a difficult and slow process due to all combinations of nanoparticles sizes, shapes, and compositions vs. biological tests, assay organisms, and multiple activity parameters to be measured. Additionally, antibiotics overuse lead to emergence of resistant strains with different metabolic networks. Computational models may speed up the process but models reported up to date do not to consider all the previous factors and data sources are disperse and not curated. In this paper, we used Information Fusion, Perturbation-Theory, and Machine Learning (IFPTML) approach, introduced by us, to fit model for antibacterial nanoparticles discovery. The dataset studied have 15 classes of nanoparticles (1-100 nm) with most cases in range 1-50 nm vs. >20 pathogenic bacteria species with different metabolic networks. The nanoparticles studied include: metal nanoparticles of Au, Ag, and Cu; oxide nanoparticles of Zn, Cu, La, Al, Fe, Sn, Ti, Cd, and Si; and metal salt nanoparticles of CuI and CdS. We used the software SOFT.PTML (our own application) with user-friendly interface for IFPTML calculations and a control statistics package. Using SOFT.PTML we found a linear logistic regression equation able to model 4 biological activity parameters using only 8 variables with 2 = 2265.75, p-level < 0.05, Sensitivity Sn = 79.4, and Specificity Sp = 99.3 for 3213 cases (nanoparticle-bacteria pairs) in training series. The model has Sn = 80.8 and Specificity Sp = 99.3 for 2114 cases in external validation series. We also seek a random forest non-linear model with higher values of Sn and Sp = 98-99% in training/validation series although more complicated to use. SOFT.PTML demonstrated to be one useful tool for the analysis of complex data in Nanotechnology. We also introduced a new Anabolism-Catabolism Unbalance index of metabolic networks to unravel the biological connotation of the IFPTML predictions for antibacterial nanoparticles. The new models open a new gate to the discovery of NPs vs. new bacterial species and strains with different topological structure of their metabolic networks.