Experimental and computational insights for broad, metagenomic next-generation sequencing-based biodetection in complex environments

Metagenomic next-generation sequencing offers a powerful approach for broadly detecting known and emerging pathogens and is being applied to an expanding set of applications from infectious disease diagnostics to biodefense. However, the ability to sequence a sample of interest does not necessarily mean that the results will be actionable for decision-makers. Generalizable, data-driven approaches are needed to quantify the performance of metagenomic workflows and enable high-confidence decision-making. Here, we generate a well-characterized matrix of experimental samples, parameterizing variables such as biological target type and quantity, background composition and complexity, and processing workflow methodology. We systematically characterize the impact these variables have on the sensitivity and specificity of target detection and benchmark performance against a gold-standard method, polymerase chain reaction. We show that the target mass fraction (the fraction of the total nucleic acid mass contributed by a target's genome) is a key determinant for signal detection, and we develop a generalizable linear mixed effects model that can be used to estimate sequencing depth requirements based on input properties. Given adequate sequencing depth, sensitivity is driven by the quantity of the target detected in the background (e.g., the Limit of Blank). We also investigate the impact that closely related organisms can have on the specificity of target detection due to mis-assignment of reads originating from shared genomic regions, especially as the ratio of target to near-neighbors decreases. Our results demonstrate that detection performance varies based on sample type, target, methodology, and use case, complicating comparison and interpretation. We therefore propose and exercise a synthetic data generation and analysis approach to quantitatively assess the sensitivity and specificity of target detection given a particular workflow, target, and sample type of interest. Collectively, our study seeks to provide insights for generalizable, data-driven approaches to characterize, optimize, and refine metagenomic workflows to enable actionable decision making.

Authors

Institutions

Publication Details

Journal
PLoS ONE
Published
2026-11-16
DOI
https://doi.org/10.1371/journal.pone.0361775
Primary Topic
Genomics and Phylogenetic Studies
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Experimental and computational insights for broad, metagenomic next-generation sequencing-based biodetection in complex environments

Amy Xiao, Meghan E. Ramsey, Jessie Hendricks, Anthony Mannion et al.
4 links
PLoS ONE
Genomics and Phylogenetic Studies
article

Experimental and computational insights for broad, metagenomic next-generation sequencing-based biodetection in complex environments

Amy Xiao, Meghan E. Ramsey, Jessie Hendricks, Anthony Mannion, Lynn Martin Kerr, Judilee D. Osborn, Jennifer L. Mallek
article en

Abstract

Metagenomic next-generation sequencing offers a powerful approach for broadly detecting known and emerging pathogens and is being applied to an expanding set of applications from infectious disease diagnostics to biodefense. However, the ability to sequence a sample of interest does not necessarily mean that the results will be actionable for decision-makers. Generalizable, data-driven approaches are needed to quantify the performance of metagenomic workflows and enable high-confidence decision-making. Here, we generate a well-characterized matrix of experimental samples, parameterizing variables such as biological target type and quantity, background composition and complexity, and processing workflow methodology. We systematically characterize the impact these variables have on the sensitivity and specificity of target detection and benchmark performance against a gold-standard method, polymerase chain reaction. We show that the target mass fraction (the fraction of the total nucleic acid mass contributed by a target's genome) is a key determinant for signal detection, and we develop a generalizable linear mixed effects model that can be used to estimate sequencing depth requirements based on input properties. Given adequate sequencing depth, sensitivity is driven by the quantity of the target detected in the background (e.g., the Limit of Blank). We also investigate the impact that closely related organisms can have on the specificity of target detection due to mis-assignment of reads originating from shared genomic regions, especially as the ratio of target to near-neighbors decreases. Our results demonstrate that detection performance varies based on sample type, target, methodology, and use case, complicating comparison and interpretation. We therefore propose and exercise a synthetic data generation and analysis approach to quantitatively assess the sensitivity and specificity of target detection given a particular workflow, target, and sample type of interest. Collectively, our study seeks to provide insights for generalizable, data-driven approaches to characterize, optimize, and refine metagenomic workflows to enable actionable decision making.

PLoS ONEVol. 21(11)
MIT Lincoln Laboratory (US), Massachusetts Institute of Technology (US)
U.S. Department of Homeland Security
Good health and well-being
Genomics and Phylogenetic Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon