Convolutional Neural Networks for Biological Sequence Taxonomic Classification: A Comparative Study
Helaly, Marwah A.; Rady, Sherine; Aref, Mostafa M.;
Abstract
Biological sequence classification is a key task in Bioinformatics. For research labs today, the classification of unknown biological sequences is essential for facilitating the identification, grouping and study of organisms and their evolution. This paper compares three of the most recent deep learning works on the 16S rRNA barcode dataset for taxonomic classification. Three different CNN architectures are compared together with three different feature representations, namely: k-mer spectral representation, Frequency Chaos Game Representation (FCGR) and character-level integer encoding. Experimental results and comparisons have shown that representations that hold positional information about the nucleotides in a sequence perform much better with accuracies reaching 91.6% on the most fine-grained classification task.
Other data
Title | Convolutional Neural Networks for Biological Sequence Taxonomic Classification: A Comparative Study | Authors | Helaly, Marwah A.; Rady, Sherine ; Aref, Mostafa M. | Keywords | Biological sequences;Deep learning;Classification;Convolutional neural networks;RNA;Feature representation;DNA | Issue Date | 1-Jan-2020 | Publisher | SPRINGER INTERNATIONAL PUBLISHING AG | Conference | Advances in Intelligent Systems and Computing | ISBN | 9783030311285 | ISSN | 21945357 | DOI | 10.1007/978-3-030-31129-2_48 | Scopus ID | 2-s2.0-85075598371 | Web of science ID | WOS:000569375900048 |
Attached Files
File | Description | Size | Format | Existing users please Login |
---|---|---|---|---|
AISI.2019_Marwa et al.pdf | 671.66 kB | Adobe PDF | Request a copy |
Similar Items from Core Recommender Database
Items in Ain Shams Scholar are protected by copyright, with all rights reserved, unless otherwise indicated.