Background: The immunotoxicity of engine exhausts is of high concern to human\nhealth due to the increasing prevalence of immune-related diseases. However, the\nevaluation of immunotoxicity of engine exhausts is currently based on expensive and\ntime-consuming experiments. It is desirable to develop efficient methods for immunotoxicity\nassessment.\nMethods: To accelerate the development of safe alternative fuels, this study proposed\na computational method for identifying informative features for predicting proinflammatory\npotentials of engine exhausts. A principal component regression (PCR)\nalgorithm was applied to develop prediction models. The informative features were\nidentified by a sequential backward feature elimination (SBFE) algorithm.\nResults: A total of 19 informative chemical and biological features were successfully\nidentified by SBFE algorithm. The informative features were utilized to develop a computational\nmethod named FS-CBM for predicting proinflammatory potentials of engine\nexhausts. FS-CBM model achieved a high performance with correlation coefficient values\nof 0.997 and 0.943 obtained from training and independent test sets, respectively.\nConclusions: The FS-CBM model was developed for predicting proinflammatory\npotentials of engine exhausts with a large improvement on prediction performance\ncompared with our previous CBM model. The proposed method could be further\napplied to construct models for bioactivities of mixtures.
Loading....