Accurate prediction and detection of the DNA regions or their underlying structural patterns are constant difficulties for researchers. Feature extraction and functional classification of genomic sequences is an inter...Accurate prediction and detection of the DNA regions or their underlying structural patterns are constant difficulties for researchers. Feature extraction and functional classification of genomic sequences is an interesting area of research. Many computational techniques have already been applied including the artificial neural network （ANN）, nonlinear model, spectrogram and statistical techniques. In this paper, some features are extracted from the wavelet coefficient and second set of features are extracted from the frequency of transition of nucleotides. These two features sets are examined. The purpose was to investigate the abilities of these parameters to predict critical segment in the DNA sequence. The neuro-fuzzy system was used for prediction. The performance of the neuro-fuzzy system was evaluated in terms of training performance and prediction accuracies. Two genomic sequences of the classes： prokaryotic and eukaryotic were used, as an example, （Escherichia coli） and （Caenorhabditis elegans） sequences were selected.展开更多
文摘Accurate prediction and detection of the DNA regions or their underlying structural patterns are constant difficulties for researchers. Feature extraction and functional classification of genomic sequences is an interesting area of research. Many computational techniques have already been applied including the artificial neural network （ANN）, nonlinear model, spectrogram and statistical techniques. In this paper, some features are extracted from the wavelet coefficient and second set of features are extracted from the frequency of transition of nucleotides. These two features sets are examined. The purpose was to investigate the abilities of these parameters to predict critical segment in the DNA sequence. The neuro-fuzzy system was used for prediction. The performance of the neuro-fuzzy system was evaluated in terms of training performance and prediction accuracies. Two genomic sequences of the classes： prokaryotic and eukaryotic were used, as an example, （Escherichia coli） and （Caenorhabditis elegans） sequences were selected.
