Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Sep;7(9):000642.
doi: 10.1099/mgen.0.000642.

Metagenomic evidence for a polymicrobial signature of sepsis

Cedric Chih Shen Tan  1   2 , Mislav Acman  1 , Lucy van Dorp  1 , Francois Balloux  1
Affiliations

Metagenomic evidence for a polymicrobial signature of sepsis

Cedric Chih Shen Tan et al. Microb Genom. 2021 Sep.

Abstract

Our understanding of the host component of sepsis has made significant progress. However, detailed study of the microorganisms causing sepsis, either as single pathogens or microbial assemblages, has received far less attention. Metagenomic data offer opportunities to characterize the microbial communities found in septic and healthy individuals. In this study we apply gradient-boosted tree classifiers and a novel computational decontamination technique built upon SHapley Additive exPlanations (SHAP) to identify microbial hallmarks which discriminate blood metagenomic samples of septic patients from that of healthy individuals. Classifiers had high performance when using the read assignments to microbial genera [area under the receiver operating characteristic (AUROC=0.995)], including after removal of species 'culture-confirmed' as the cause of sepsis through clinical testing (AUROC=0.915). Models trained on single genera were inferior to those employing a polymicrobial model and we identified multiple co-occurring bacterial genera absent from healthy controls. While prevailing diagnostic paradigms seek to identify single pathogens, our results point to the involvement of a polymicrobial community in sepsis. We demonstrate the importance of the microbial component in characterising sepsis, which may offer new biological insights into the aetiology of sepsis, and ultimately support the development of clinical diagnostic or even prognostic tools.

Keywords: SHAP; bacteraemia; blood metagenomics; contamination; kitome; machine learning; metagenomics; sepsis.

PubMed Disclaimer

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Figures

Fig. 1.
Fig. 1.
Model interpretation and performance. (a) Plot summarizing the SHAP values across all samples for the most important features ranked by the mean absolute SHAP value (highest at the top) for Karius-Neat, (b) Karius-SD, (c) Karius-CR and (d) Karius-Without models. Each point represents a single sample. Points with similar SHAP values were stacked vertically for visualization of point density and were coloured according to the magnitude of the feature values (i.e. read counts). Genera that contained ‘culture-confirmed’ pathogens are highlighted in yellow.
Fig. 2.
Fig. 2.
Comparison of performance (AUROC) for the multi-feature models (Karius-Neat, Karius-Only, Karius-Without feature space) and single-feature models (x-axis). Models were optimised and evaluated using the nested cross-validation protocol.
Fig. 3.
Fig. 3.
Generalisability of models across sepsis cohorts. Model performance before and after SHAP Decontamination determined via holdout cross-validation (see Methods). The table appended describes the sepsis definition used, sequencing type and test size for each holdout dataset, and the corresponding size of the training data.
Fig. 4.
Fig. 4.
Corrected microbial co-occurrence network for genera assigned in sepsis metagenomes. Input data correspond to the Karius-SD feature space. The edges in this network represent those in the septic network that were not present in the healthy network. The widths of edges are weighted by the strength of the SparCC correlations. Nodes are coloured as per the legend at the top, with ‘culture-confirmed’ pathogens those experimentally shown to be implicated in sepsis. The layout of the graph was generated using the Fruchterman–Reingold algorithm.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Rudd KE, Johnson SC, Agesa KM, Shackelford KA, Tsoi D, et al. Global, regional, and national sepsis incidence and mortality, 1990–2017: analysis for the Global Burden of Disease Study. Lancet. 2020;395:200–211. doi: 10.1016/S0140-6736(19)32989-7. - DOI - PMC - PubMed
    1. Kwizera A, Baelani I, Mer M, Kissoon N, Schultz MJ, et al. The long sepsis journey in low-and middle-income countries begins with a first step… but on which road? Springer. 2018 - PMC - PubMed
    1. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest. 1992;101:1644–1655. doi: 10.1378/chest.101.6.1644. - DOI - PubMed
    1. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, et al. 2001 sccm/esicm/accp/ats/sis international sepsis definitions conference. Intensive Care Med. 2003;29:530–538. doi: 10.1007/s00134-003-1662-x. - DOI - PubMed
    1. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3. Jama. 2016;315:801–810. doi: 10.1001/jama.2016.0287. - DOI - PMC - PubMed

Publication types