Meta-analysis on plant-mediated synthesized gold and silver nanoparticles

Abstract

Gold and silver nanoparticles (Au and Ag-NPs) have found much broader applications owing to their high surface-to-volume ratio, which makes them easy to penetrate the surface and interact with bio-fluids. This has raised an important concern over the toxicity of these NPs. The literature says that though the plant-mediated synthesis of Au and Ag-NPs is environment-friendly and more economical, yet some show cytotoxic characteristics. It has also been noted that Au and Ag NPs are promising materials in health care and medical devices. Ourobjective is to comprehend the parameters that make NPs cytotoxic or non-cytotoxic. We seek to investigate the effect of the plant-mediated synthesis on the physical properties, viz., morphology, surface plasmon resonance, size, stability etc., through machine learning algorithms like DT, RF and k-means clustering. The parameters used for machine learning models included the physical characteristics and biosynthesized related variables that affect the cytotoxicity of NPs. Through the assertion of specific threshold values, DT and RF helped to categorize the input parameters. The precise prediction and classification of toxicity and non-toxicity on normal and cancer cell lines were supported by the regression matrices. The high R2 (DT-0.8, RF-0.84) and relatively low RMSE (DT-13.54, RF-12.29) values as it is in relevance the dataset and for the model it is evaluated and illustrated in table 1. The comparison and correlation among the input features were also studied through a scatter plot of k-means clusters.