Several recent works have shown that K-mer sequence representation of a DNA sequence can be used for classification or identification of nucleosome positioning related sequences. This representation can be computationally expensive when k grows, making the complexity in spaces of exponential dimension. This issue effects significantly the classification task computed by a general machine learning algorithm used for the purpose of sequence classification. In this paper, we investigate the advantage offered by the so-called Variable Ranking Feature Selection method to select the most informative k − mers associated to a set of DNA sequences, for the final purpose of nucleosome/linker classification by a deep learning network. Results computed on three public datasets show the effectiveness of the adopted feature selection method.
|Titolo della pubblicazione ospite||New Trends in Databases and Information Systems|
|Numero di pagine||11|
|Stato di pubblicazione||Published - 2018|
|Nome||COMMUNICATIONS IN COMPUTER AND INFORMATION SCIENCE|