Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Improved prediction of bacterial transcription start sites
Bioinformatics
Expert Systems with Applications: An International Journal
Refinement of approximate domain theories by knowledge-based neural networks
AAAI'90 Proceedings of the eighth National conference on Artificial intelligence - Volume 2
Hi-index | 12.05 |
A novel strategy to the prediction of promoter sequences, which was based on the analysis of chemical features, was developed for the first time. A string of nucleotides was translated into numerical sequences by means of the chemical parameters that represented the chemical properties and molecular structures of nucleotides, and then genetic algorithm was employed to select effective chemical features so as to establish the proper predictive model of support vector machine (SVM). The accuracies of the final SVM model for the leave-one-out cross-validation on the training set were 100%, and the sensitivity and specificity reached to 1. The accuracy for testing set was also up to 100%. Moreover, several functional sites and chemical parameters selected by SVM model were discussed. The satisfactory results indicated that the study of chemical features in sequences was effective, and the proposed approach was reliable.