Monitoring of Antimicrobial Resistance Based on Metagenomics Analyses in Pneumonia Patients

Overview

Monitoring of antimicrobial resistance (AMR) based on metagenomics analyses in pneumonia patients is critical for optimizing clinical diagnosis and treatment and improving clinical prognosis. This study is designed to ask the following key questions: 1. What is the microbiome maps of patients with severe pneumonia and mild pneumonia ? 2. How many pathogen resistance genes are carrying in severe pneumonia and mild pneumonia ? 3. What is the genetic diversity of key pathogens detected in severe pneumonia and mild pneumonia during 2019-2025?

This is a historical prospective obsevational study. Patients diagnosed with severe and mild pneumonia are recruited continously from four hospitals (Shanghai General Hospital, Wuxi People's Hospital Affiliated to Nanjing Medical University, Wuhan Union Hospital, and Huanggang Central Hospital) in China during March, 2019 to March, 2025. Next-generation sequencing: Metagenomics and metatranscriptomic libraries undergo next-generation sequencing using the Illumina Novaseq 6000 platform. Microbiome analysis: using KneadData (v0.10.0) reference from the human genome (GRCH38 reference database) to filter out the quality control of illumina sequencing data in Virosaurus download virus genome data sets for reference, bowtie2 (v2.3.4.1) (genome coverage \>30%, depth \>1X) was used to locate and analyze the post-host sequence, and then the "samtools idxstats" command was used to calculate the classification and relative abundance of viruses. Meanwhile, in order to obtain the annotation information of bacteria at the species level, PhyloFlash (v3.4) was used to calculate read counts for 16S rRNA genes in the SILVA database, selecting similarity greater than or equal to 98% as a threshold. Using eukaryotic pathogen genome database EUPATHDB46 as reference, bowtie2 and samtools were used for qualitative and quantitative analysis of fungal pathogens. The criteria for determining the cause of respiratory infection are: (1) existing species known to be associated with human disease (ICD-10), (2) previously unidentified potential novel pathogens (only DNA and RNA viruses whose genera or families have previously been shown to infect mammals), and (3) possible symbiotic bacteria not included. Analysis of AMR: by comparing the sequence similarity between the sequencing fragments and known drug resistance genes, the detection content can determine whether drug resistance genes exist, and suggest drug resistance caused by modification, inactivation, repression and other drug resistance genes. Drug resistance genes detection: genes related to drug resistance recorded in CARD (Comprehensive Antibiotic Resistance Database) and ARG-ANNOT database. In this assay, only functional genes with drug resistance activities such as modification, inactivation, and repression, as well as pathway and target changes caused by some point mutations, were reported. Genetic diversity was computed as the mean pairwise genetic distance within a group. Maximum likelihood phylogenetic trees were constructed using RaxML with a general time-reversible nucleotide substitution model and 1000 bootstraps. The genetic distance between sequences was calculated using MEGAX, with a bootstrap method for variance estimation.