Download This Paper
Preprints with The Lancet is part of SSRN´s First Look, a place where journals identify content of interest prior to publication. Authors have opted in at submission to The Lancet family of journals to post their preprints on Preprints with The Lancet. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early stage research papers that have not been peer-reviewed. The findings should not be used for clinical or public health decision making and should not be presented to a lay audience without highlighting that they are preliminary and have not been peer-reviewed. For more information on this collaboration, see the comments published in The Lancet about the trial period, and our decision to make this a permanent offering, or visit The Lancet´s FAQ page, and for any feedback please contact preprints@lancet.com.
Integrating Respiratory Microbiome and Host Immune Response Using Machine Learning for Diagnosis of the Lower Respiratory Tract Infections
28 Pages Posted: 13 Jul 2023
More...Abstract
Background: At present, the diagnosis of lower respiratory tract infections (LRTIs) is difficult and there is an urgent need for better diagnostic methods.
Methods: This study enrolled 136 patients from 2020 to 2021 and collected bronchoalveolar lavage fluid (BALF) specimens. We used metatranscriptome to analyze the lower respiratory tract microbiome and host immune response.
Findings: The diversity of the lower respiratory tract microbiota in LRTIs significantly decreased, manifested by a decrease in the abundance of normal microbiota in the oropharynx and an increase in the abundance of opportunistic pathogenic bacteria. The upregulated differentially expressed genes (DEGs) in the LRTIs group were mainly enriched in infection immune response-related pathways. Klebsiella pneumoniae had the most significant increase in abundance in LRTIs, which was associated with upregulation of the host gene CMTM1, which belongs to the chemokine-like factor family. We combined clinical information, lower respiratory tract microbiome, and host transcriptome data to construct a machine learning model with 70 screened features to predict LRTIs. The results showed that the model trained by random forest in the validation set had the best performance (ROC AUC: 0.957). The independent external dataset showed an accuracy of 88.2% for this model.
Interpretation: This study suggests that the prediction model integrating clinical information, lower respiratory tract microbiome, and host transcriptome data can be an effective tool for LRTIs diagnosis.
Funding: This work was supported by the National Key R&D Program of China (2022YFA1304300) funded to H.C., and Beijing Municipal Science and Technology Commission program (Z191100006619100) funded to H.C..
Declaration of Interest: All of authors declare that they have no competing interests.
Ethical Approval: This study was approved by the research ethics board at Peking University People’s Hospital (approval no. 2019PHB134).
Keywords: lower respiratory tract infection, pneumonia, next-generation sequencing, transcriptome, machine learning
Suggested Citation: Suggested Citation