PAIPline: pathogen identification in metagenomic and clinical next generation sequencing samples

Andrusch, Andreas; Dabrowski, Piotr Wojtek; Klenner, Jeanette; Tausch, Simon H.; Kohl, Claudia; Osman, Abdalla A.; Renard, Bernhard Y.; Nitsche, Andreas

doi:10.1093/bioinformatics/bty595

Cited by 29 publications

(20 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[30] Technological advances in NGS have also allowed for rapid pathogen identification during outbreaks and pathogen discovery. [31–33] The current landscape with regards to these technological frontiers suggests that it may take substantial time and investment before these technologies achieve levels of performance, ease-of-use, and affordability required in order to become widely available and accessible across the globe.…”

Section: Discussionmentioning

confidence: 99%

Global knowledge gaps in acute febrile illness etiologic investigations: A scoping review

et al. 2019

View full text Add to dashboard Cite

BackgroundAcute febrile illness (AFI), a common reason for people seeking medical care globally, represents a spectrum of infectious disease etiologies with important variations geographically and by population. There is no standardized approach to conducting AFI etiologic investigations, limiting interpretation of data in a global context. We conducted a scoping review to characterize current AFI research methodologies, identify global research gaps, and provide methodological research standardization recommendations.Methodology/FindingsUsing pre-defined terms, we searched Medline, Embase, and Global Health, for publications from January 1, 2005–December 31, 2017. Publications cited in previously published systematic reviews and an online study repository of non-malarial febrile illness etiologies were also included. We screened abstracts for publications reporting on human infectious disease, aimed at determining AFI etiology using laboratory diagnostics. One-hundred ninety publications underwent full-text review, using a standardized tool to collect data on study characteristics, methodology, and laboratory diagnostics. AFI case definitions between publications varied: use of self-reported fever as part of case definitions (28%, 53/190), fever cut-off value (38·0°C most commonly used: 45%, 85/190), and fever measurement site (axillary most commonly used: 19%, 36/190). Eighty-nine publications (47%) did not include exclusion criteria, and inclusion criteria in 13% (24/190) of publications did not include age group. No publications included study settings in Southern Africa, Micronesia & Polynesia, or Central Asia. We summarized standardized reporting practices, specific to AFI etiologic investigations that would increase inter-study comparability.ConclusionsWider implementation of standardized AFI reporting methods, with multi-pathogen disease detection, could improve comparability of study findings, knowledge of the range of AFI etiologies, and their contributions to the global AFI burden. These steps can guide resource allocation, strengthen outbreak detection and response, target prevention efforts, and improve clinical care, especially in resource-limited settings where disease control often relies on empiric treatment. PROSPERO: CRD42016035666.

show abstract

Section: Discussionmentioning

confidence: 99%

Global knowledge gaps in acute febrile illness etiologic investigations: A scoping review

et al. 2019

View full text Add to dashboard Cite

show abstract

“…As the state-of-the art approach for the open-view detection of pathogens is genome sequencing ( 5 , 6 ), it is crucial to develop automated pipelines for characterizing the infectious potential of currently unidentifiable sequences. In practice, clinical samples are dominated by host reads and contaminants, with often less than a hundred reads of the pathogenic virus ( 7 ). Metagenomic assembly is challenging, especially in time-critical applications.…”

Section: Introductionmentioning

confidence: 99%

Interpretable detection of novel human viruses from genome sequencing data

Bartoszewicz

Seidel

Renard

2021

NAR Genomics and Bioinformatics

Self Cite

View full text Add to dashboard Cite

Viruses evolve extremely quickly, so reliable methods for viral host prediction are necessary to safeguard biosecurity and biosafety alike. Novel human-infecting viruses are difficult to detect with standard bioinformatics workflows. Here, we predict whether a virus can infect humans directly from next-generation sequencing reads. We show that deep neural architectures significantly outperform both shallow machine learning and standard, homology-based algorithms, cutting the error rates in half and generalizing to taxonomic units distant from those presented during training. Further, we develop a suite of interpretability tools and show that it can be applied also to other models beyond the host prediction task. We propose a new approach for convolutional filter visualization to disentangle the information content of each nucleotide from its contribution to the final classification decision. Nucleotide-resolution maps of the learned associations between pathogen genomes and the infectious phenotype can be used to detect regions of interest in novel agents, for example, the SARS-CoV-2 coronavirus, unknown before it caused a COVID-19 pandemic in 2020. All methods presented here are implemented as easy-to-install packages not only enabling analysis of NGS datasets without requiring any deep learning skills, but also allowing advanced users to easily train and explain new models for genomics.

show abstract

“…As the state-of-the art approach for the openview detection of pathogens is genome sequencing (5,6), it is crucial to develop automated pipelines for characterizing the infectious potential of currently unidentifiable sequences. In practice, clinical samples are dominated by host reads and contaminants, with often less than a hundred reads of the pathogenic virus (7). Metagenomic assembly is challenging, especially in time-critical applications.…”

Section: Introductionmentioning

confidence: 99%

Interpretable detection of novel human viruses from genome sequencing data

Bartoszewicz

Seidel

Renard

2020

Preprint

View full text Add to dashboard Cite

AbstractViruses evolve extremely quickly, so reliable methods for viral host prediction are necessary to safeguard biosecurity and biosafety alike. Novel human-infecting viruses are difficult to detect with standard bioinformatics workflows. Here, we predict whether a virus can infect humans directly from next-generation sequencing reads. We show that deep neural architectures significantly outperform both shallow machine learning and standard, homology-based algorithms, cutting the error rates in half and generalizing to taxonomic units distant from those presented during training. We propose a new approach for convolutional filter visualization to disentangle the information content of each nucleotide from its contribution to the final classification decision. Nucleotide-resolution maps of the learned associations between pathogen genomes and the infectious phenotype can be used to detect virulence-related genes in novel agents, for example the SARS-CoV-2 coronavirus, unknown before it caused a COVID-19 pandemic in 2020.

show abstract

PAIPline: pathogen identification in metagenomic and clinical next generation sequencing samples

Cited by 29 publications

References 25 publications

Global knowledge gaps in acute febrile illness etiologic investigations: A scoping review

Global knowledge gaps in acute febrile illness etiologic investigations: A scoping review

Interpretable detection of novel human viruses from genome sequencing data

Interpretable detection of novel human viruses from genome sequencing data

Contact Info

Product

Resources

About