The promises and pitfalls of machine learning for detecting viruses in aquatic metagenomes

Alise J. Ponsero, Bonnie L Hurwitz

Research output: Contribution to journalReview article

Abstract

Tools allowing for the identification of viral sequences in host-associated and environmental metagenomes allows for a better understanding of the genetics and ecology of viruses and their hosts. Recently, new approaches using machine learning methods to distinguish viral from bacterial signal using k-mer sequence signatures were published for identifying viral contigs in metagenomes. The promise of these content-based approaches is the ability to discover new viruses, with no or few known relatives. In this perspective paper, we examine the use of the content-based machine learning tool VirFinder for the identification of viral sequences in aquatic metagenomes and explore the possibility of using ecosystem-focused models targeted to marine metagenomes. We discuss the impact of the training set composition on the tool performance and the current limitation for the retrieval of low abundance viral sequences in metagenomes. We identify potential biases that could arise from machine learning approaches for viral hunting in real-world datasets and suggest possible avenues to overcome them.

Original languageEnglish (US)
Article number806
JournalFrontiers in Microbiology
Volume10
Issue numberMAR
DOIs
StatePublished - Jan 1 2019

Fingerprint

Metagenome
Viruses
Ecology
Ecosystem
Machine Learning

Keywords

  • Machine learning
  • Metagenomic
  • Sequence classification
  • Viral signature
  • Virus

ASJC Scopus subject areas

  • Microbiology
  • Microbiology (medical)

Cite this

The promises and pitfalls of machine learning for detecting viruses in aquatic metagenomes. / Ponsero, Alise J.; Hurwitz, Bonnie L.

In: Frontiers in Microbiology, Vol. 10, No. MAR, 806, 01.01.2019.

Research output: Contribution to journalReview article

@article{85c6bb54c18549898c160040497fdbb7,
title = "The promises and pitfalls of machine learning for detecting viruses in aquatic metagenomes",
abstract = "Tools allowing for the identification of viral sequences in host-associated and environmental metagenomes allows for a better understanding of the genetics and ecology of viruses and their hosts. Recently, new approaches using machine learning methods to distinguish viral from bacterial signal using k-mer sequence signatures were published for identifying viral contigs in metagenomes. The promise of these content-based approaches is the ability to discover new viruses, with no or few known relatives. In this perspective paper, we examine the use of the content-based machine learning tool VirFinder for the identification of viral sequences in aquatic metagenomes and explore the possibility of using ecosystem-focused models targeted to marine metagenomes. We discuss the impact of the training set composition on the tool performance and the current limitation for the retrieval of low abundance viral sequences in metagenomes. We identify potential biases that could arise from machine learning approaches for viral hunting in real-world datasets and suggest possible avenues to overcome them.",
keywords = "Machine learning, Metagenomic, Sequence classification, Viral signature, Virus",
author = "Ponsero, {Alise J.} and Hurwitz, {Bonnie L}",
year = "2019",
month = "1",
day = "1",
doi = "10.3389/fmicb.2019.00806",
language = "English (US)",
volume = "10",
journal = "Frontiers in Microbiology",
issn = "1664-302X",
publisher = "Frontiers Media S. A.",
number = "MAR",

}

TY - JOUR

T1 - The promises and pitfalls of machine learning for detecting viruses in aquatic metagenomes

AU - Ponsero, Alise J.

AU - Hurwitz, Bonnie L

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Tools allowing for the identification of viral sequences in host-associated and environmental metagenomes allows for a better understanding of the genetics and ecology of viruses and their hosts. Recently, new approaches using machine learning methods to distinguish viral from bacterial signal using k-mer sequence signatures were published for identifying viral contigs in metagenomes. The promise of these content-based approaches is the ability to discover new viruses, with no or few known relatives. In this perspective paper, we examine the use of the content-based machine learning tool VirFinder for the identification of viral sequences in aquatic metagenomes and explore the possibility of using ecosystem-focused models targeted to marine metagenomes. We discuss the impact of the training set composition on the tool performance and the current limitation for the retrieval of low abundance viral sequences in metagenomes. We identify potential biases that could arise from machine learning approaches for viral hunting in real-world datasets and suggest possible avenues to overcome them.

AB - Tools allowing for the identification of viral sequences in host-associated and environmental metagenomes allows for a better understanding of the genetics and ecology of viruses and their hosts. Recently, new approaches using machine learning methods to distinguish viral from bacterial signal using k-mer sequence signatures were published for identifying viral contigs in metagenomes. The promise of these content-based approaches is the ability to discover new viruses, with no or few known relatives. In this perspective paper, we examine the use of the content-based machine learning tool VirFinder for the identification of viral sequences in aquatic metagenomes and explore the possibility of using ecosystem-focused models targeted to marine metagenomes. We discuss the impact of the training set composition on the tool performance and the current limitation for the retrieval of low abundance viral sequences in metagenomes. We identify potential biases that could arise from machine learning approaches for viral hunting in real-world datasets and suggest possible avenues to overcome them.

KW - Machine learning

KW - Metagenomic

KW - Sequence classification

KW - Viral signature

KW - Virus

UR - http://www.scopus.com/inward/record.url?scp=85066444217&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066444217&partnerID=8YFLogxK

U2 - 10.3389/fmicb.2019.00806

DO - 10.3389/fmicb.2019.00806

M3 - Review article

AN - SCOPUS:85066444217

VL - 10

JO - Frontiers in Microbiology

JF - Frontiers in Microbiology

SN - 1664-302X

IS - MAR

M1 - 806

ER -