(510b) Association Rule Mining of the Relationships Among Biological Responses of Embryonic Zebrafish Exposed to Nanoparticles

Relationships among the biological responses of embryonic zebrafish (including mortality) exposed to seven different types of nanoparticles at multiple concentrations were analyzed using association rule mining and clustering (hierarchical clustering and self-organizing maps (SOM)) analysis. Biological response data from a database on nanomaterial biological interactions (NBI) were extracted and used for clustering analysis and subsequently processed (categorized) for the application of association rule mining. The dataset consisted of ENMs characteristics, experimental conditions, ENMs concentrations, and effects of nanoparticles on embryonic zebrafish. Association rule mining was then applied to the biological responses dataset, followed by the identification of non-redundant rules. The non-redundant rules indicated that a significant impact of ENMs exposure to zebrafish on one or more biological responses implies significant impact of ENMs exposure to zebrafish on other (associated) responses. Pearson’s pairwise correlation analysis, along with hierarchical clustering and SOM analysis indicated that the non-redundant association rules were consistent with confirmatory tests. The experimental results confirmed that zebrafish olfactory regions are inter-associated and heart is inter-associated with several responses especially touch response, circulation, delayed development, yolk sac edema, curved body axis, caudal fin, and pectoral fin deformations. Spontaneous zebrafish movements also correlated with yolk sac edema, axis and somites. The results of the current study suggest that there is merit in exploring the utility of the approach to streamline multiparametric biological responses for improved understanding of their relationships for the development of predictive toxicity nano-structure activity relationships.