Development of an Artificial Neural Network Model for Risk Assessment of Skin Sensitization Using Human Cell Line Activation Test, Direct Peptide Reactivity Assay, KeratinoSens™ and in Silico Structure Alert Parameter

摘要: It is important to predict the potential of cosmetic ingredients cause skin sensitization, and in accordance with European Union directive for replacement animal tests, several vitro tests based on adverse outcome pathway have been developed hazard identification, such as direct peptide reactivity assay, KeratinoSens™ human cell line activation test. Here, we describe development an artificial neural network (ANN) prediction model sensitization risk assessment integrated testing strategy concept, using KeratinoSens™, test silico or structure alert parameter. We first investigated relationship between published murine local lymph node assay EC3 values, which represent potency, results a panel about 134 chemicals all required data were available. Predictions ANN analysis combinations parameters from three showed good correlation values. However, when was applied set 28 that had not included training set, predicted EC3s overestimated some chemicals. Incorporation additional descriptor (obtained TIMES-M Toxtree software) improved results. Our findings suggest concept could be useful evaluating potential.