Publication detail

Reference-free Identification of Phage DNA Using Signal Processing on Nanopore Data

KUPKOVÁ, K. SEDLÁŘ, K. PROVAZNÍK, I.

Original Title

Reference-free Identification of Phage DNA Using Signal Processing on Nanopore Data

Type

conference paper

Language

English

Original Abstract

Nanopore sequencing has become an invaluable aid in small sequencing projects. Thanks to its compact size, the Oxford Nanopore MinION platform is often used in crisis situations, such as outbreaks of microbial infections, to determine the causes of the problem. As a platform that produces data in real-time, it requires bioinformatics techniques designed for fast data processing. In this paper, we demonstrate the possibility of the direct processing of nanopore current signals, the so-called squiggles, for fast reference-free identification of phage DNA. The proposed technique is based on the computation of Hjorth parameters and is suitable for fast visualization of the data, as well as for proper classification by many machine learning algorithms. The classification of the data also raises the possibility of applying adapted base calling algorithms for both groups separately, as phage and host DNA have different features.

Keywords

nanopore sequencing; shotgun; signal processing; phage; SVM; AdaBoost; random forrest

Authors

KUPKOVÁ, K.; SEDLÁŘ, K.; PROVAZNÍK, I.

Released

23. 10. 2017

Publisher

Conference Publishing Services

Location

Washington DC

ISBN

978-1-5386-1324-5

Book

2017 IEEE 17th International Conference on Bioinformatics and Bioengineering, BIBE 2017

Pages from

101

Pages to

105

Pages count

584

BibTex

@inproceedings{BUT140931,
  author="Kristýna {Kupková} and Karel {Sedlář} and Valentine {Provazník}",
  title="Reference-free Identification of Phage DNA Using Signal Processing on Nanopore
Data",
  booktitle="2017 IEEE 17th International Conference on Bioinformatics and Bioengineering, BIBE 2017",
  year="2017",
  pages="101--105",
  publisher="Conference Publishing Services",
  address="Washington DC",
  doi="10.1109/BIBE.2017.00-71",
  isbn="978-1-5386-1324-5"
}