Next Article in Journal
Changes in Soil Microbial Community Structure and Assembly Process Under Different Forest Restoration Strategies in Cold Temperate Forests of Northeastern China
Next Article in Special Issue
The Identification of Novel Mutations in ATP-Dependent Protease ClpC1 Assists in the Molecular Diagnosis of Obscured Pyrazinamide-Resistant Tuberculosis Clinical Isolates
Previous Article in Journal
Co-Inoculating Burkholderia vietnamiensis B418 and Trichoderma harzianum T11W Reduced Meloidogyne incognita Infestation of Tomato Plants
Previous Article in Special Issue
Comparative Assessment of Rapid Identification and Antimicrobial Susceptibility Testing Methods for Bloodstream Infections in a Non-24/7 Clinical Microbiology Laboratory
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Microorganisms
Volume 13
Issue 6
10.3390/microorganisms13061338
microorganisms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Shotgun Metagenomic Sequencing Analysis as a Diagnostic Strategy for Patients with Lower Respiratory Tract Infections

by
Ha-eun Cho
1,†,
Min Jin Kim
2,†,
Jongmun Choi
2,
Yong-Hak Sohn
2,
Jae Joon Lee
2,
Kyung Sun Park
3,
Sun Young Cho
3,
Ki-Ho Park
4 and
Young Jin Kim
3,*
1
Department of Laboratory Medicine, Graduate School, Kyung Hee University, Kyung Hee University Hospital, Seoul 02447, Republic of Korea
2
Department of Laboratory Medicine, Seegene Medical Foundation, Seoul 04805, Republic of Korea
3
Department of Laboratory Medicine, Kyung Hee University College of Medicine, Kyung Hee University Hospital, Seoul 02447, Republic of Korea
4
Department of Infectious Diseases, Kyung Hee University College of Medicine, Kyung Hee University Hospital, Seoul 02447, Republic of Korea
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Microorganisms 2025, 13(6), 1338; https://doi.org/10.3390/microorganisms13061338
Submission received: 12 May 2025 / Revised: 5 June 2025 / Accepted: 7 June 2025 / Published: 9 June 2025
(This article belongs to the Special Issue Detection and Analysis of Clinical Microbial Infections, Second Edition)
Browse Figure
Review Reports Versions Notes

Abstract

Conventional diagnostic methods (CDMs) for lower respiratory infections (LRIs) have limitations in detecting causative pathogens. This study evaluates the utility of shotgun metagenomic sequencing (SMS) as a complementary diagnostic tool using bronchoalveolar lavage (BAL) fluid. Sixteen BAL fluid samples from pneumonia patients with positive CDM results—including bacterial/fungal cultures; PCR for Mycobacterium tuberculosis or cytomegalovirus; and the BioFire® FilmArray® Pneumonia Panel (BioFire Diagnostics LLC, Salt Lake City, UT, USA)—underwent 10 Gb SMS on the Illumina NovaSeq 6000 platform (Illumina, San Diego, CA, USA). Reads were aligned to the NCBI RefSeq database; with fungal identification further supported by internal transcribed spacer (ITS) analysis. Antibiotic resistance genes (ARGs) were annotated using the Comprehensive Antibiotic Resistance Database. Microbial reads accounted for 0.00002–0.04971% per sample. SMS detected corresponding bacteria in 63% of cases, increasing to 69% when subdominant taxa were included. Fungal reads were low; however, Candida species were identified in four samples via ITS. No viral reads were detected. ARGs meeting perfect match criteria were found in two cases. This is the first real-world study comparing SMS with CDMs, including semiquantitative PCR, in BAL fluid for LRI. SMS shows promise as a supplementary diagnostic method, with further research needed to optimize its performance and cost-effectiveness.

1. Introduction

Identifying the causative pathogen of infectious pneumonia is crucial for targeted treatment and improved patient outcomes [1]. The detection rate of pathogens in patients with lower respiratory infections (LRIs) ranges from 38–46% [2,3]. Despite the identification of multiple microorganisms using conventional diagnostic methods (CDMs), a single pathogen is typically considered the primary cause of infection [4]. Although polymicrobial infections are reported in 5.7–38.4% of LRI cases, their etiology is rarely confirmed during treatment [5,6,7,8,9].
Bacterial etiology accounts for over 50% of diagnoses, leading to the prioritization of empirical antibiotic use [10]. However, the spectrum of causative microbes is diverse [11]. Culture and polymerase chain reaction (PCR) are standard diagnostic tools supplemented by antigen tests. However, cultures may fail to detect fastidious bacteria and fungi [12,13], while targeted PCR may overlook microbes not included in the assay [14]. Multiplex PCR panels expand the detection range of clinically relevant microbes, enabling comprehensive identification. Nevertheless, these PCR panels have limitations, particularly when pathogens not covered in the panel proliferate and contribute to infection [15]. Antigen tests provide a rapid and cost-effective diagnostic option but are hindered by low sensitivity and false positives [13]. The variability in pathogen detection rates across these diagnostic methods highlights the complexity of achieving accurate LRI diagnoses, emphasizing the need for complementary diagnostic approaches.
In response, shotgun metagenomic sequencing (SMS) has gained attention. This approach involves analyzing all nucleic acids within a sample, enabling the comprehensive identification of potential pathogens. Furthermore, the use of extensive reference databases enhances the potential for syndromic testing, offering universal pathogen detection [16]. Studies utilizing SMS on bronchoalveolar lavage (BAL) specimens have demonstrated a sensitivity of 88–97% and a specificity of 15–81% for accurate pathogen identification [17,18,19,20].
Owing to its anatomical location, BAL fluid inherently possesses low microbial biomass [21], posing challenges for microbial signal detection. Therefore, optimizing methodologies is crucial for achieving reliable SMS results. Despite the development of semiquantitative multiplex PCR, few studies have compared its performance to that of SMS. Furthermore, discussions on the appropriate capacity for SMS of BAL fluid remain limited. The absence of consensus on the criteria for interpreting SMS positivity further complicates the issue [17,18,19,20,22,23,24,25,26].
This study presents the first comparative assessment of SMS and conventional diagnostic tests in clinical practice, including semiquantitative multiplex PCR. The performance of SMS with a 10 Gb output was evaluated, along with efforts to optimize the criteria for SMS interpretation.

2. Materials and Methods

2.1. Sample Collection and Processing

From March to July 2023, a total of 44 BAL fluid samples with positive results were consecutively collected from the BioFire® FilmArray® Pneumonia Panel (FA-PP; BioFire Diagnostics LLC, Salt Lake City, UT, USA). The FA-PP is a semi-quantitative PCR test representing the latest diagnostic method for pathogen detection. Exclusion criteria for sample selection were as follows: (1) clinical diagnosis of non-infectious pneumonia, (2) contamination with normal flora [27], and (3) final diagnosis of pneumonia caused by RNA viruses. Initially, 12 cases diagnosed with non-infectious diseases were excluded, leaving 32 cases for further analysis (Figure 1).
CDMs were performed using standard clinical protocols for pathogen detection, including bacterial, fungal, and Mycobacterium tuberculosis/non-tuberculous mycobacteria (MTB/NTM) cultures. Additional tests included matrix-assisted laser desorption/ionization time-of-flight mass spectrometry system (MADLI-TOF MS; Bruker Daltonics, Billerica, MA, USA), TB/NTM PCR (AdvanSure™ TB/NTM real-time PCR kit on AdvanSure SLAN 96 and E3 system; LG Life Science, Seoul, Korea), Xpert MTB/RIF assay (Cepheid, Sunnyvale, CA, USA), cytomegalovirus (CMV) PCR (nucleic acid extracted using QiaAmp DSP DNA mini kit on QIAcube; Qiagen GmbH, Hilden, Germany, and PCR performed on CFX 96; Bio-Rad, Hercules, CA, USA) and Pneumocystis jirovecii PCR (nucleic acid extracted using a laboratory-developed test reagent on MagNa Pure 96; Roche Diagnostics, Mannheim, Germany, and PCR performed on CFX 96; Bio-Rad, Hercules, CA, USA). Antigen detection tests included PLATELIA Aspergillus Ag (Bio-Rad, Hercules, CA, USA), targeting galactomannan of Aspergillus spp., and Pastorex Crypto Plus (Sanofi-Diagnostics Pasteur, Marnes-La-Coquette, France), detecting cryptococcal capsular polysaccharide glucuronoxylomannan (Table S1). Antimicrobial susceptibility testing (AST) for cultured isolates was performed using the MicroScan WalkAway 96 plus system (Beckman Coulter, Brea, CA, USA) with NM44, PM28, and MSTRP+1 panels to determine minimum inhibitory concentrations (MICs). The positive reporting criterion for microbial growth was defined as greater than 104 CFU/mL in culture [27]. Species identification was assigned to isolates with MALDI-TOF MS scores ≥ 2.0 [28,29]. FA-PP positivity was defined as exceeding 104 copies/mL. TB/NTM PCR targeted the IS6110 region specific to the MTB complex and the internal transcribed spacer (ITS) region of mycobacteria, using a Ct cutoff of 35. The Xpert MTB/RIF assay detected IS1081 and IS6110 sequences and assessed rifampin resistance based on rpoB target probes. CMV PCR employed a TaqMan probe labeled with VIC and MGBNFQ, with Ct < 40 indicating positive amplification. P. jirovecii PCR was interpreted using a Ct threshold of 35.
Based on culture results, 14 samples with a high risk of contamination were additionally excluded, following these criteria:
(1)
Specimens with a high presence of oropharyngeal normal flora, such as Streptococcus mitis, Streptococcus australis, Streptococcus parasanguinis, Streptococcus sanguinis, Streptococcus clone, Streotpcoccus gordonii, Streptococcus intermedius, Gemella haemolysans, Gemella sanguinis, Granulicatella adiacens, Granulicatella elegans, Abiotrophia defective, and Rothia dentocariosa, were excluded. These microorganisms were identified and presumed to be contaminants during the BAL procedure [30,31,32].
(2)
Specimens with the identification of cutaneous normal flora, such as Cutibacteriium acnes, Cutibacterium granulosum, Corynebacterium striatum, Staphylococcus epidermidis, Staphylococcus hominis, Staphylococcus haemolyticus, Staphylococcus captis, Staphylococcus warneri, Staphylococcus saprophyticus, Staphylococcus cohnnii, Staphylococcus xylosus, Staphylococcus simulans, Micrococcus luteus, and Micrococcus varians, were excluded. These microorganisms were identified and presumed to pose a risk of contamination during specimen collection and processing [31,33,34,35,36].
(3)
Commensal microorganisms, including Staphylococcus aureus, were identified in the study. Despite their classification as commensals, their potential to cause pneumonia was considered significant [37,38].
These samples included samples containing normal oropharyngeal flora, which may have been introduced during the BAL procedure [32], and normal cutaneous flora, which could have led to contamination during specimen collection and processing [36]. However, commensal microorganisms with a high potential to cause pneumonia, such as Staphylococcus aureus, as well as cases in which pathogens such as TB were co-detected, were included in the study [37].
During preliminary nucleic acid extraction experiments conducted for this study, RNA did not meet the minimum quality threshold (RIN ≥ 5), whereas DNA consistently showed DIN ≥ 3, allowing stable library preparation. Therefore, SMS was performed using DNA. Consequently, two FA-PP–positive samples that were clinically diagnosed as infectious pneumonia but yielded only RNA viruses as pathogens were excluded. Ultimately, 16 samples were included in the final analysis.

2.2. Extraction of Nucleic Acids and Sequencing

DNA was extracted from 16 BAL fluid samples using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). The extracted nucleic acid concentration was measured using the Qubit dsDNA HS Assay Kit on a Qubit 4.0 instrument (Life Technologies, Carlsbad, CA, USA) and the Nanodrop One (Thermo Fisher Scientific, Waltham, MA, USA). DNA integrity was assessed using the Genomic DNA ScreenTape Assay Kit on a TapeStation 4200 System (Agilent Technologies, Inc., Santa Clara, CA, USA).
Libraries were constructed using Illumina DNA Prep (Illumina, San Diego, CA, USA). Quality control was performed using the TapeStation 4200 System with the D1000 ScreenTape Assay Kit (Agilent Technologies, Inc.).
Shotgun sequencing was performed using a paired-end configuration with 150 bp × 2 reads on the NovaSeq 6000 instrument (Illumina, San Diego, CA, USA), utilizing SP Flow Cell reagents and BCL Convert.

2.3. SMS Procedures

The sequencing output was set to a capacity of 10 Gb. Following BCL Convert processing, FastQC software (v0.12.0) was used to assess raw data quality by evaluating the proportion of reads with a quality score ≥ Q30 and the GC content. Adapter sequences were largely removed by BCL Convert with high accuracy. Therefore, no additional trimming was performed to avoid excessive data loss, considering both data yield and quality.
Raw reads were mapped to the human reference genome (GRCh38.p14) using URMAP and Samtools, and unmapped reads were extracted and merged using BBMap. Taxonomic assignment of the reads was performed using MMseqs2 against the SILVA_138 database, applying an initial e-value cutoff of 1 × 10−5 to identify unique genes across ecosystems. Parameters for taxonomic classification included a minimum alignment length of 90 bp and a minimum identity of 80%, with the best alignment prioritized.
All unmapped reads were assembled using SPAdes with the metaviral option enabled to recover scaffolds corresponding to viral genomes. Reference viral genomes relevant to LRIs were included, such as adenovirus, CMV, Epstein–Barr virus, varicella-zoster virus, and herpes simplex virus [3,11,39,40]. Viral reads were matched to reference viral genomes using MMseqs2 for taxonomic annotation [41].
For further taxonomic classification of shotgun reads, both small subunit (SSU) and internal transcribed spacer (ITS) regions were aligned against the SILVA_138 database using MMseqs2 and analyzed using STAMP [42]. Broader microbial classification was performed using microbial genome databases from the NCBI RefSeq database (release 13 October 2023). Bacterial sequences not classifiable to the genus or species level were labeled as unclassified. Fungal sequences were analyzed using ITS targets, especially in samples with confirmed fungal growth or PCR positivity.
To profile the resistome, antibiotic resistance genes (ARGs) were identified using Resistance Gene Identifier (RGI) software (v6.0.3) and the CARD database (v3.2.8). Identified genes were annotated based on the CARD model. The “perfect” algorithm of RGI, which detects 100% matches to curated reference sequences and known resistance-conferring mutations, was used as the default threshold [42].
Criteria for pathogen detection by SMS reads were established based on prior literature. For bacterial detection, a relative abundance of ≥30% was considered positive [17,18,20,26]. MTB was considered positive with ≥1 mapped read [17,18,19,22,23,24,26], while NTM were considered positive if ranked among the top 10 most abundant bacterial taxa [23,24]. For fungal detection, a species was considered positive when its coverage was at least fivefold higher than that of any other fungal species [22,23,26]. In cases with low coverage, fungal reads were considered positive if alignment length was ≥100 bp and identity ≥ 98%. Viral detection was defined as ≥3 mapped reads [18,24,43].
Microbial reads that met these thresholds were classified as SMS-positive. Subdominant microbes below the threshold were also reviewed. The microbe deemed clinically relevant by the attending physician—based on treatment decisions—was considered the causative agent.
ARG results were compared with AST results from cultured isolates. If the AST confirmed resistance to antibiotics associated with the identified ARGs, the result was deemed consistent. If the isolate showed susceptibility or indeterminate results to any antibiotics in the relevant class, the finding was considered inconsistent. Only “perfect hits”—defined as sequences with 100% identity to reference ARGs—were used as the threshold in this study, based on prior evidence showing 89–95% concordance with culture-based AST; “strict hits” (≥80% identity) showed lower concordance (65%) and were not used [44,45,46].

3. Results

3.1. Identification of Microbes by SMS

Among the 16 FA-PP-positive samples, the number of reads mapped to the human genome ranged from 34,553,153 to 55,464,668. Microbial reads were detected in all cases, comprising 0.00002–0.04971% of the total reads. Of these, 99.3% were bacterial reads matched to the 16S rRNA gene, 0.3% were eukaryotic reads matched to the 18S rRNA gene, and 0.4% were fungal reads matched to the ITS gene.
SMS detected pathogens above the threshold in 10 of 16 cases (63%), increasing to 11 of 16 cases (69%) when subdominant taxa were included. CDMs identified bacterial pathogens in 11 of 16 cases (69%), including two cases of MTB. Among the 16 cases, bacterial–fungal co-infections were observed in 2 of 16 cases (12.5%), a bacterial–viral co-infection in 1 case (6.25%), and a fungal–viral co-infection in 1 case (6.25%). In the 11 cases in which CDMs identified bacteria as the causative pathogen, SMS detected the corresponding bacteria in nine cases. These included Pseudomonas aeruginosa in cases 3, 12, 14, and 15; Haemophilus influenzae in cases 10 and 11; Klebsiella pneumoniae in case 7; Acinetobacter baumannii in case 13; and Stenotrophomonas maltophilia in case 16. Expanding the criteria to include subdominant bacterial reads, K. pneumoniae was additionally detected in case 8, increasing the detection rate to 10 out of 11 bacterial cases. In cases where fungi were the primary pathogen, SMS detected Candida tropicalis reads in case 4 (Table 1).

3.2. Metagenomic Results of Antibiotic Resistance

Results meeting the perfect criteria of the RGI were observed in cases 2 and 14.
In case 2, K. pneumoniae was evaluated against AST results. The sul1 gene, associated with sulfonamide resistance, was detected and corresponded with resistance to trimethoprim/sulfamethoxazole in AST. No carbapenem-related ARGs were identified, which was consistent with AST results showing susceptibility to doripenem, ertapenem, imipenem, and meropenem. However, AAC(6′)-Ia, associated with aminoglycoside resistance, was detected but was inconsistent with AST results, which showed susceptibility to amikacin and gentamicin (Table 2).
In case 14, P. aeruginosa infection was evaluated. MexI and H-NS, which are associated with fluoroquinolone and tetracycline resistance, were detected and were consistent with AST results showing resistance to levofloxacin and tetracycline. However, several inconsistencies were observed. OXA-217, smeR, H-NS, and CTX-M-15, which are associated with resistance to penicillin derivatives and cephalosporins, were detected. However, AST results showed resistance only to ampicillin, piperacillin, and cefotaxime, while susceptibility was observed for cefepime and ceftazidime. For carbapenem resistance, AXC-1 and OXA-217 were identified; however, AST results indicated susceptibility to doripenem, imipenem, and meropenem. Additionally, AAC(6′)-Ia, cpxA, and smeR, linked to aminoglycoside resistance, were detected, whereas AST results showed susceptibility to amikacin and an indeterminate result for tobramycin (Table 3).
Regarding antibiotic resistance, no bacteriophages were detected in Case 2, whereas Inoviridae species were identified in Case 14.

4. Discussion

The proportion of microbial reads averaged 0.00412% of the total reads, ranging from 0.00002–0.04971% per sample. This aligns with the recommended range of 0.00001–0.7% for successful SMS application in clinical settings [47]. Previous studies have established absolute criteria for bacterial positivity in SMS (Table 4), typically defining a relative abundance threshold of ≥30% [17,18,20,26]. However, some cases exhibited SMS-detected reads for subdominant taxa that matched the primary pathogen identified by CDMs. For example, in case 8, K. pneumoniae was identified with a relative abundance of 17.78% but was classified as positive using adjusted criteria that considered subdominant taxa. While FA-PP reported K. pneumoniae at 105 copies/mL, culture results indicated normal respiratory flora (NRF), which was insufficient to establish it as the primary pathogen. This finding highlights the need for caution when interpreting taxa with relative abundances below 30%, as their classification as pathogens requires additional corroborative evidence. Without clear supporting data, taxa with low relative abundances should not be automatically assumed to represent the primary pathogen. However, their potential clinical relevance warrants careful consideration depending on the clinical context.
Despite applying expanded criteria, SMS did not detect the pathogen in 31% of cases (5 out of 16). This included four cases (involving MTB, Aspergillus sp., or CMV) in which no reads were detected, and one case in which FA-PP identified K. pneumoniae at 105 copies/mL; however, no corresponding reads were found in SMS.
Although MTB was detected in cases 1 and 2 by CDMs, SMS did not identify any MTB reads. This contrasts with previous studies that report a sensitivity of 44–48% for MTB detection by SMS, comparable to PCR [48,49]. Some studies have reported higher SMS sensitivity than target-specific real-time PCR [50,51]; however, these studies included diverse specimen types, such as lung biopsy tissue, limiting direct comparisons. Several factors may account for the lower sensitivity observed in this study. First, the nucleic acid extraction method used for SMS may have been suboptimal for mycobacteria. A study comparing sputum DNA extraction kits for Mycobacterium spp. demonstrated significant differences in 16S rRNA gene cycle threshold (Ct) values depending on the kit used [52]. While no direct comparison exists between the QIAamp DNA Mini Kit used in this study and those used in previous SMS studies, kit-related variations may have influenced the results. The robust, waxy cell wall of Mycobacterium spp. makes lysis challenging, often leading to low DNA yield [53]. Additionally, a study on sputum lysis methods for Mycobacterium spp. reported differences in Ct value standard deviation based on lysis temperature, further emphasizing the impact of extraction conditions [54]. Second, differences in sample preprocessing may have affected SMS sensitivity. Variations in centrifugal force and duration have been shown to influence Mycobacterium sedimentation, affecting recovery rates and smear sensitivity [55]. Notably, our study applied a higher relative centrifugal force than previous studies that reported greater SMS sensitivity for MTB. Excessive centrifugal force can generate heat, potentially leading to bacterial injury and reduced DNA recovery [49,50,55]. The absence of phenol treatment, which has been suggested to enhance DNA purity, may have further contributed to reduced detection [53,56]. Finally, the patients’ treatment history may have influenced SMS sensitivity. Previous studies have reported a significant decrease in MTB detection sensitivity from 76% in pretreatment samples to 31% in post-treatment samples [49]. Both cases 1 and 2 involved patients with a history of anti-tuberculosis treatment, and neither case exhibited rifampin resistance. In case 2, M. tuberculosis was detected by the Xpert MTB/RIF assay and cultured; however, Corynebacterium striatum was the predominant isolate, suggesting that competition with other organisms and prior treatment may have reduced SMS sensitivity. Enhancing MTB detection may require developing targeted panels optimized for Mycobacterium specimen processing.
In case 6, in which Aspergillus sp. was detected by CDMs, SMS did not yield any reads, likely due to the incomplete or variable nature of ITS reference sequences, which complicates fungal identification via SMS [57]. In this study, fungal species identified via CDMs were not detected as positive reads using SMS. The ITS1 sequence is recognized for its reliability in identifying Candida spp., Pneumocystis spp., and Aspergillus spp. within the Ascomycota division, offering accurate species- and genus-level classification [58]. Although fungal pneumonia is increasingly prevalent among immunocompromised individuals, all cases in this study were HIV-negative [13]. Notably, P. jirovecii pneumonia detection has been increasing among HIV-uninfected immunocompromised individuals, highlighting its clinical relevance [59]. However, SMS did not detect P. jirovecii reads in this study, even with the complete ITS1 sequence. Previous studies have reported significantly lower quantities of Pneumocystis spp. in BAL fluid from HIV-negative patients than from HIV-positive patients, which may have contributed to the challenges in SMS-based detection [60].
For cases 5, 6, and 16, CMV was considered the primary pathogen based on clinical assessment, with PCR results ranging from 9665 to 200,175 copies/mL. In case 4, CMV was detected at 1,343,600 copies/mL; however, it was not targeted for treatment owing to the patient’s immunological status. These findings highlight the limitations of relying solely on CDMs for clinical decision-making, as CMV PCR results require integration with a broader clinical context. Notably, SMS did not detect viral reads in any case, precluding the identification of CMV or other viruses. Despite DNA extraction yields exceeding 3–8 times the 4 nM threshold for library preparation, the proportion of valid viral sequences remained low (~0.05%), indicating potential biases [61]. The small size of viral genomes further complicates SMS-based detection [62]. Although the DNA integrity number exceeded the manufacturer’s recommended threshold of 3, no standardized criteria exist for library construction. Additionally, while the Q30 score surpassed the 85% threshold, it did not reach 90%, which may have affected viral read recovery [63]. Moreover, extended storage and delayed processing for SMS, compared to immediate testing for FA-PP and CMV PCR, may have contributed to viral degradation and reduced detection rates [64]. Bacteriophages, which constitute the majority of viral particles in both environmental and human-associated microbiomes, account for over 90% of the human virome [65,66,67]. Incorporating bacteriophage detection strategies into viral metagenomic analyses may enhance SMS-based viral identification.
All bacteria detected as positive by SMS in this study were Gram-negative. Gram-negative bacilli are the most common causative agents of LRIs in elderly patients [4], and the 10 Gb SMS platform demonstrated strong performance in detecting these pathogens. In Case 2, where Corynebacterium striatum was the predominant isolate by culture, SMS revealed Serratia spp. as the second most abundant taxon. Among these, reads corresponding to S. marcescens (26 reads, 6.25%)—a known cause of LRI in hospitalized patients—were detected, along with S. rubidaea (4 reads, 1%). S. rubidaea is a rare but emerging pathogen capable of causing LRIs, as reported in several case studies [68]. Due to its overlapping biochemical characteristics with lactose-fermenting Enterobacterales, S. rubidaea is difficult to identify using CDMs [68,69]. This case illustrates how SMS can aid in the detection of potentially pathogenic bacteria that are difficult to isolate or identify via culture due to sample-related or organism-specific limitations and the 10 Gb SMS demonstrated strong performance in detecting these pathogens.
Unlike previous studies, this study incorporated FA-PP into CDMs, potentially enhancing the pathogen detection rate. FA-PP has been shown to identify target pathogens even in samples reported as “no growth” or “normal flora” in culture [70]. However, no clear correlation was observed between FA-PP semi-quantitative values and SMS read numbers. For FA-PP values of 105 copies/mL, SMS reads ranged from 5–52; for 106 copies/mL, reads ranged from 2–310; and for values exceeding 107 copies/mL, reads ranged from 347–23,869. A culture result of 104 CFU/mL corresponded to 105 copies/mL in FA-PP, 3000–6000 CFU/mL corresponded to 106 copies/mL, and ≥105 CFU/mL aligned with 106–107 copies/mL in FA-PP.
This study partially investigated genotypic AST using SMS in BAL fluid; however, the results did not consistently correlate with phenotypic AST of cultured isolates. Using the CARD database, SMS-detected ARGs were compared with phenotypic AST results from CDM-identified pathogens [44]. In previous studies, the use of the perfect threshold showed a discordance rate of only 8%, significantly lower than the 35% observed with the strict threshold, suggesting that the perfect criterion reduces false positives and improves clinical reliability [45,46]. The strict threshold has been criticized for potentially identifying non-functional genes unrelated to resistance phenotypes, which may mislead treatment decisions. Conversely, the perfect threshold is considered more suitable for providing actionable insights for targeted therapy, as it minimizes overprediction of resistome diversity and accounts better for clinical relevance [45,46]
Despite applying only the perfect algorithm in this study, ARGs could not be conclusively linked to the cultured pathogens. In case 14, Inoviridae sp. was detected, suggesting that bacteriophage-mediated mechanisms may have influenced resistance expression or contributed to indeterminate AST results, despite the presence of ARGs. Although AXC-1 and OXA-217 were identified, phenotypic carbapenem resistance was not observed. This may be attributed to Inoviridae sp. enhancing P. aeruginosa biofilm formation and stability, thereby reducing antibiotic penetration [71,72]. Additionally, the indeterminate result for aminoglycosides may be partially explained by reduced outer membrane permeability associated with cpxA regulation influenced by Inoviridae sp. [73]. The high severity of illness in this case, with a pneumonia severity index (PSI) of 164 (grade 5), also raises the possibility that bacterial virulence was enhanced by bacteriophage interaction [74].
A key limitation of SMS is the inability to trace specific ARGs to their bacterial hosts [75], which may lead to an overestimation of resistance, especially when resistance is confined to a single antibiotic within a class. Although ARGs are detectable at the DNA level, their functional expression at the RNA or protein level remains unknown, limiting insights into actual resistance mechanisms. A meta-analysis of genotypic AST using metagenomic sequencing reported a categorical agreement of 88%; however, very major errors (VME) (24%, 95% CI: 8–40%) and major errors (ME) (5%, 95% CI: 0–12%) exceeded the US Food and Drug Administration (FDA)-recommended thresholds [75]. While a machine learning-based genotypic AST model met FDA requirements (ME ≤ 3%, VME ≤ 1.5%) with high performance [76], clinical specimens pose additional challenges. The presence of host nucleic acids, multiple pathogens, and plasmid-mediated ARGs complicates species attribution [76,77,78]. Additionally, database limitations and low area under the curve (AUC) values in some models highlight the need for further research to refine genotypic AST approaches [76].
However, the direct application of SMS-detected ARGs in clinical practice remains premature. While SMS cannot conclusively determine the direct involvement of ARGs in resistance phenotypes, it enables the detection of microbes at very low concentrations and provides a comprehensive overview of the resistome within a sample [70,75]. The administration of antibiotics to patients colonized with multidrug-resistant organisms increases colonization density and expands the resistance gene pool, thereby increasing the risk of infection [76,79]. The maintenance of resistance-conferring mutations in bacterial pathogens is entirely dependent on their effect on fitness and virulence [80]. Given the association between the resistome and microbial diversity, antimicrobial stewardship is increasingly emphasized. Overcoming current limitations may allow SMS-based ARG identification to contribute to these efforts [76].
This study demonstrates the feasibility of achieving sufficient sequencing coverage for SMS with a 10 Gb output. Previous studies have not provided detailed sequencing configurations for BAL fluid. While highly complex samples such as stool require 1–10 Gb (with >7 Gb recommended), lower-complexity samples such as anterior nares may be adequately analyzed with <1 Gb [81,82]. To assess whether 10 Gb was sufficient for BAL fluid analysis, we conducted experiments based on existing recommendations [47]. Increasing sequencing output generally enhances pathogen detection; however, in untargeted approaches without host genome depletion, a proportional increase in host-derived sequences may limit improvements in sensitivity [83]. Using a 10 Gb configuration, DNA-based SMS detected primary pathogens in 63% of cases based on applied thresholds, increasing to 69% when subdominant taxa were included. In SMS, the presence of host nucleic acid affects sensitivity, and appropriate depletion can increase the relative abundance of microbial signals [84]. However, unintentional removal of microbial content along with host DNA and the variability of results depending on sample type remains controversial [85]. In this study, SMS was conducted without host genome depletion. Further research is required to validate the effective host signal removal in BAL samples.
Inconsistencies between SMS and clinical diagnoses are well-documented, with comparative studies reporting low agreement levels (κ = 0.035–0.347) between SMS and CDM positivity [22,23,26]. In this study, cases with negative SMS results despite positive CDM findings underscore the limitations of SMS as a molecular diagnostic tool [86]. Challenges include differentiating viable pathogens from residual nucleic acids in resolved or treated infections and distinguishing asymptomatic colonization from true infection [14]. Additionally, variations in bacterial read distributions complicate the application of uniform positivity thresholds. Future studies should refine SMS interpretation criteria by integrating clinical features to improve their correlation with clinical outcomes. Comparative analyses of sequencing output capacities for BAL fluid samples are required to determine the optimal conditions for pathogen detection. Further research should also explore methodologies that expand SMS applications beyond the bacteriome to include the mycobiome and virome, thereby enhancing its diagnostic potential.

5. Conclusions

This study presents a comparison between SMS and CDMs, elucidating the performance and limitations of SMS with a 10 Gb output for LRI diagnosis. While SMS cannot replace PCR, it can complement culture methods, serving a mutually supportive role. To establish SMS as a valuable supplementary tool in LRI diagnostics, further research is needed to explore its variable applications for higher sensitivity and cost-effectiveness.
Abbreviations: CDM, conventional diagnostic method; BAL, bronchoalveolar lavage; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry; FA-PP, FilmArray® Pneumonia Panel; MTB, Mycobacterium tuberculosis; NTM, non-tuberculous mycobacteria; MTB, Mycobacterium tuberculosis; CMV, cytomegalovirus.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/microorganisms13061338/s1. Table S1. Performance of CDMs for BAL fluids. References [87,88,89,90,91,92,93,94,95,96] are cited in the Supplementary Materials.

Author Contributions

Conceptualization, H.-e.C. and Y.J.K.; Data curation, H.-e.C., M.J.K., J.J.L., K.S.P. and S.Y.C.; Formal analysis, H.-e.C., K.-H.P. and Y.J.K.; Investigation, H.-e.C. and M.J.K.; Methodology, M.J.K., J.C. and Y.-H.S.; Visualization, H.-e.C. and Y.J.K.; Supervision, Y.J.K.; Writing—Original draft preparation, H.-e.C.; Writing—Review and editing: Y.J.K.; Funding acquisition, K.-H.P. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by a grant from the National Research Foundation of Korea, funded by the Korean government (MSIT) (No. RS-2023-00246999). Ki-Ho Park received the funding. The funders played no role in the study design, data collection, data analysis, data interpretation, or manuscript writing.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board (IRB) of Kyung Hee University Medical Center (protocol code: 2023-05-076, date of approval: 1 June 2023). The approved collection period for archived residual samples extended until 29 February 2024.

Informed Consent Statement

Informed consent was waived by the IRB, as the study utilized archived residual and anonymized BAL fluid samples originally collected for quality control purposes. All samples were de-identified prior to analysis to ensure anonymity. The residual samples were securely stored and managed according to the Bioethics and Safety Act and were disposed of following the Waste Management Act after the completion of the study. Research-related documents will be entrusted to the IRB document custodian and preserved for three years, after which they will be securely destroyed.

Data Availability Statement

The raw sequencing data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) under the accession number PRJNA1036216.

Acknowledgments

We express our profound gratitude to the laboratory and medical personnel for their diligent execution of clinical measurements, whose contributions have been instrumental in facilitating the success of this project. We extend our appreciation to the reviewers for their valuable suggestions and insights. We convey our gratitude to all the authors for their dedicated efforts toward the culmination of this project.

Conflicts of Interest

The authors declare no conflict of interest. The sponsor had no role in the design, execution, interpretation, or writing of the study.

References

  1. Amati, F.; Bindo, F.; Stainer, A.; Gramegna, A.; Mantero, M.; Nigro, M.; Bussini, L.; Bartoletti, M.; Blasi, F.; Aliberti, S. Identify Drug-Resistant Pathogens in Patients with Community-Acquired Pneumonia. Adv. Respir. Med. 2023, 91, 224–238. [Google Scholar] [CrossRef] [PubMed]
  2. Chen, S.Y.; Chen, Y.J.; Er, T.K. Retrospective Study of Lower Respiratory Tract Infections in the Intensive Care Unit Detected by the FilmArray Pneumonia Panel. Clin. Lab. 2023, 69. Available online: https://pubmed.ncbi.nlm.nih.gov/37436371/ (accessed on 6 June 2025). [CrossRef] [PubMed]
  3. Jain, S.; Self, W.H.; Wunderink, R.G.; Fakhran, S.; Balk, R.; Bramley, A.M.; Reed, C.; Grijalva, C.G.; Anderson, E.J.; Courtney, D.M.; et al. Community-Acquired Pneumonia Requiring Hospitalization among U.S. Adults. N. Engl. J. Med. 2015, 373, 415–427. [Google Scholar] [CrossRef] [PubMed]
  4. Tarsia, P.; Aliberti, S.; Pappalettera, M.; Blasi, F. Mixed community-acquired lower respiratory tract infections. Curr. Infect. Dis. Rep. 2007, 9, 14–20. [Google Scholar] [CrossRef]
  5. Cilloniz, C.; Ewig, S.; Ferrer, M.; Polverino, E.; Gabarrus, A.; Puig de la Bellacasa, J.; Mensa, J.; Torres, A. Community-acquired polymicrobial pneumonia in the intensive care unit: Aetiology and prognosis. Crit. Care 2011, 15, R209. [Google Scholar] [CrossRef]
  6. de Roux, A.; Ewig, S.; Garcia, E.; Marcos, M.A.; Mensa, J.; Lode, H.; Torres, A. Mixed community-acquired pneumonia in hospitalised patients. Eur. Respir. J. 2006, 27, 795–800. [Google Scholar] [CrossRef]
  7. Gutierrez, F.; Masia, M.; Rodriguez, J.C.; Mirete, C.; Soldan, B.; Padilla, S.; Hernandez, I.; Royo, G.; Martin-Hidalgo, A. Community-acquired pneumonia of mixed etiology: Prevalence, clinical characteristics, and outcome. Eur. J. Clin. Microbiol. Infect. Dis. 2005, 24, 377–383. [Google Scholar] [CrossRef]
  8. Holter, J.C.; Muller, F.; Bjorang, O.; Samdal, H.H.; Marthinsen, J.B.; Jenum, P.A.; Ueland, T.; Froland, S.S.; Aukrust, P.; Husebye, E.; et al. Etiology of community-acquired pneumonia and diagnostic yields of microbiological methods: A 3-year prospective study in Norway. BMC Infect. Dis. 2015, 15, 64. [Google Scholar] [CrossRef]
  9. Lieberman, D.; Schlaeffer, F.; Boldur, I.; Lieberman, D.; Horowitz, S.; Friedman, M.G.; Leiononen, M.; Horovitz, O.; Manor, E.; Porath, A. Multiple pathogens in adult patients admitted with community-acquired pneumonia: A one year prospective study of 346 consecutive patients. Thorax 1996, 51, 179–184. [Google Scholar] [CrossRef]
  10. GBD 2016 Lower Respiratory Infections Collaborators. Estimates of the global, regional, and national morbidity, mortality, and aetiologies of lower respiratory infections in 195 countries, 1990-2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Infect. Dis. 2018, 18, 1191–1210. [Google Scholar] [CrossRef]
  11. Carroll, K.C.; Adams, L.L. Lower Respiratory Tract Infections. Microbiol. Spectr. 2016, 4. Available online: https://pubmed.ncbi.nlm.nih.gov/27726814/ (accessed on 6 June 2025). [CrossRef] [PubMed]
  12. Doern, G.V. Detection of selected fastidious bacteria. Clin. Infect. Dis. 2000, 30, 166–173. [Google Scholar] [CrossRef]
  13. Kelly, B.T.; Pennington, K.M.; Limper, A.H. Advances in the diagnosis of fungal pneumonias. Expert Rev. Respir. Med. 2020, 14, 703–714. [Google Scholar] [CrossRef]
  14. Schmitz, J.E.; Stratton, C.W.; Persing, D.H.; Tang, Y.W. Forty Years of Molecular Diagnostics for Infectious Diseases. J. Clin. Microbiol. 2022, 60, e0244621. [Google Scholar] [CrossRef] [PubMed]
  15. Lee, S.H.; Ruan, S.Y.; Pan, S.C.; Lee, T.F.; Chien, J.Y.; Hsueh, P.R. Performance of a multiplex PCR pneumonia panel for the identification of respiratory pathogens and the main determinants of resistance from the lower respiratory tract specimens of adult patients in intensive care units. J. Microbiol. Immunol. Infect. 2019, 52, 920–928. [Google Scholar] [CrossRef]
  16. Schlaberg, R.; Chiu, C.Y.; Miller, S.; Procop, G.W.; Weinstock, G.; Professional Practice Committee and Committee on Laboratory Practices of the American Society for Microbiology; Microbiology Resource Committee of the College of American Pathologists. Validation of Metagenomic Next-Generation Sequencing Tests for Universal Pathogen Detection. Arch. Pathol. Lab. Med. 2017, 141, 776–786. [Google Scholar] [CrossRef]
  17. Chen, Y.; Feng, W.; Ye, K.; Guo, L.; Xia, H.; Guan, Y.; Chai, L.; Shi, W.; Zhai, C.; Wang, J.; et al. Application of Metagenomic Next-Generation Sequencing in the Diagnosis of Pulmonary Infectious Pathogens From Bronchoalveolar Lavage Samples. Front. Cell. Infect. Microbiol. 2021, 11, 541092. [Google Scholar] [CrossRef] [PubMed]
  18. Huang, J.; Jiang, E.; Yang, D.; Wei, J.; Zhao, M.; Feng, J.; Cao, J. Metagenomic Next-Generation Sequencing versus Traditional Pathogen Detection in the Diagnosis of Peripheral Pulmonary Infectious Lesions. Infect. Drug Resist. 2020, 13, 567–576. [Google Scholar] [CrossRef]
  19. Wang, J.; Han, Y.; Feng, J. Metagenomic next-generation sequencing for mixed pulmonary infection diagnosis. BMC Pulm. Med. 2019, 19, 252. [Google Scholar] [CrossRef]
  20. Zhang, X.; Qin, Y.; Lei, W.; Huang, J.A. Metagenomic next-generation sequencing of BALF for the clinical diagnosis of severe community-acquired pneumonia in immunocompromised patients: A single-center study. Exp. Ther. Med. 2023, 25, 178. [Google Scholar] [CrossRef]
  21. Saladie, M.; Caparros-Martin, J.A.; Agudelo-Romero, P.; Wark, P.A.B.; Stick, S.M.; O’Gara, F. Microbiomic Analysis on Low Abundant Respiratory Biomass Samples; Improved Recovery of Microbial DNA From Bronchoalveolar Lavage Fluid. Front. Microbiol. 2020, 11, 572504. [Google Scholar] [CrossRef] [PubMed]
  22. Li, J.; Zhou, C.E.; Wei, S.C.; Wang, L.N.; Shi, M.W.; Sun, C.P.; Lin, L.J.; Liu, X.M. Diagnostic Value of Metagenomic Next-Generation Sequencing for Pneumonia in Immunocompromised Patients. Can. J. Infect. Dis. Med. Microbiol. 2022, 2022, 5884568. [Google Scholar] [CrossRef]
  23. Lin, P.; Chen, Y.; Su, S.; Nan, W.; Zhou, L.; Zhou, Y.; Li, Y. Diagnostic value of metagenomic next-generation sequencing of bronchoalveolar lavage fluid for the diagnosis of suspected pneumonia in immunocompromised patients. BMC Infect. Dis. 2022, 22, 416. [Google Scholar] [CrossRef] [PubMed]
  24. Lin, T.; Tu, X.; Zhao, J.; Huang, L.; Dai, X.; Chen, X.; Xu, Y.; Li, W.; Wang, Y.; Lou, J.; et al. Microbiological diagnostic performance of metagenomic next-generation sequencing compared with conventional culture for patients with community-acquired pneumonia. Front. Cell. Infect. Microbiol. 2023, 13, 1136588. [Google Scholar] [CrossRef]
  25. Liu, H.; Zhang, Y.; Chen, G.; Sun, S.; Wang, J.; Chen, F.; Liu, C.; Zhuang, Q. Diagnostic Significance of Metagenomic Next-Generation Sequencing for Community-Acquired Pneumonia in Southern China. Front. Med. 2022, 9, 807174. [Google Scholar] [CrossRef]
  26. Sun, T.; Wu, X.; Cai, Y.; Zhai, T.; Huang, L.; Zhang, Y.; Zhan, Q. Metagenomic Next-Generation Sequencing for Pathogenic Diagnosis and Antibiotic Management of Severe Community-Acquired Pneumonia in Immunocompromised Adults. Front. Cell. Infect. Microbiol. 2021, 11, 661589. [Google Scholar] [CrossRef]
  27. Leber, A.L. Clinical Microbiology Procedures Handbook; American Society for Microbiology Press: New York, NY, USA, 2023. [Google Scholar]
  28. Schulthess, B.; Bloemberg, G.V.; Zbinden, A.; Mouttet, F.; Zbinden, R.; Bottger, E.C.; Hombach, M. Evaluation of the Bruker MALDI Biotyper for Identification of Fastidious Gram-Negative Rods. J. Clin. Microbiol. 2016, 54, 543–548. [Google Scholar] [CrossRef]
  29. Schulthess, B.; Bloemberg, G.V.; Zbinden, R.; Bottger, E.C.; Hombach, M. Evaluation of the Bruker MALDI Biotyper for identification of Gram-positive rods: Development of a diagnostic algorithm for the clinical laboratory. J. Clin. Microbiol. 2014, 52, 1089–1097. [Google Scholar] [CrossRef]
  30. Aas, J.A.; Paster, B.J.; Stokes, L.N.; Olsen, I.; Dewhirst, F.E. Defining the normal bacterial flora of the oral cavity. J. Clin. Microbiol. 2005, 43, 5721–5732. [Google Scholar] [CrossRef]
  31. Davis, C.P. Normal Flora. In Medical Microbiology, 4th ed.; Baron, S., Ed.; University of Texas Medical Branch at Galveston: Galveston, TX, USA, 1996. Available online: https://pubmed.ncbi.nlm.nih.gov/21413249/ (accessed on 6 June 2025).
  32. Pang, J.A.; Cheng, A.F.; Chan, H.S.; French, G.L. Special precautions reduce oropharyngeal contamination in bronchoalveolar lavage for bacteriologic studies. Lung 1989, 167, 261–267. [Google Scholar] [CrossRef]
  33. Chiller, K.; Selkin, B.A.; Murakawa, G.J. Skin microflora and bacterial infections of the skin. J. Investig. Dermatol. Symp. Proc. 2001, 6, 170–174. [Google Scholar] [CrossRef] [PubMed]
  34. Mushtaq, A.; Chen, D.J.; Strand, G.J.; Dylla, B.L.; Cole, N.C.; Mandrekar, J.; Patel, R. Clinical significance of coryneform Gram-positive rods from blood identified by MALDI-TOF mass spectrometry and their susceptibility profiles—A retrospective chart review. Diagn. Microbiol. Infect. Dis. 2016, 85, 372–376. [Google Scholar] [CrossRef]
  35. Noble, W.C. Skin flora of the normal and immune compromised host. Curr. Probl. Dermatol. 1989, 18, 37–41. [Google Scholar] [CrossRef]
  36. Roth, R.R.; James, W.D. Microbiology of the skin: Resident flora, ecology, infection. J. Am. Acad. Dermatol. 1989, 20, 367–390. [Google Scholar] [CrossRef]
  37. Lambotte, O.; Timsit, J.F.; Garrouste-Orgeas, M.; Misset, B.; Benali, A.; Carlet, J. The significance of distal bronchial samples with commensals in ventilator-associated pneumonia: Colonizer or pathogen? Chest 2002, 122, 1389–1399. [Google Scholar] [CrossRef]
  38. Rasmussen, T.R.; Korsgaard, J.; Moller, J.K.; Sommer, T.; Kilian, M. Quantitative culture of bronchoalveolar lavage fluid in community-acquired lower respiratory tract infections. Respir. Med. 2001, 95, 885–890. [Google Scholar] [CrossRef] [PubMed]
  39. Mohsen, A.H.; McKendrick, M. Varicella pneumonia in adults. Eur. Respir. J. 2003, 21, 886–891. [Google Scholar] [CrossRef]
  40. Simoons-Smit, A.M.; Kraan, E.M.; Beishuizen, A.; Strack van Schijndel, R.J.; Vandenbroucke-Grauls, C.M. Herpes simplex virus type 1 and respiratory disease in critically-ill patients: Real pathogen or innocent bystander? Clin. Microbiol. Infect. 2006, 12, 1050–1059. [Google Scholar] [CrossRef] [PubMed]
  41. Mirdita, M.; Steinegger, M.; Breitwieser, F.; Soding, J.; Levy Karin, E. Fast and sensitive taxonomic assignment to metagenomic contigs. Bioinformatics 2021, 37, 3029–3031. [Google Scholar] [CrossRef]
  42. Parks, D.H.; Tyson, G.W.; Hugenholtz, P.; Beiko, R.G. STAMP: Statistical analysis of taxonomic and functional profiles. Bioinformatics 2014, 30, 3123–3124. [Google Scholar] [CrossRef]
  43. Wang, H.; Lu, Z.; Bao, Y.; Yang, Y.; de Groot, R.; Dai, W.; de Jonge, M.I.; Zheng, Y. Clinical diagnostic application of metagenomic next-generation sequencing in children with severe nonresponding pneumonia. PLoS ONE 2020, 15, e0232610. [Google Scholar] [CrossRef] [PubMed]
  44. Alcock, B.P.; Huynh, W.; Chalil, R.; Smith, K.W.; Raphenya, A.R.; Wlodarski, M.A.; Edalatmand, A.; Petkau, A.; Syed, S.A.; Tsang, K.K.; et al. CARD 2023: Expanded curation, support for machine learning, and resistome prediction at the Comprehensive Antibiotic Resistance Database. Nucleic Acids Res. 2023, 51, D690–D699. [Google Scholar] [CrossRef]
  45. Arango-Argoty, G.; Garner, E.; Pruden, A.; Heath, L.S.; Vikesland, P.; Zhang, L. DeepARG: A deep learning approach for predicting antibiotic resistance genes from metagenomic data. Microbiome 2018, 6, 23. [Google Scholar] [CrossRef]
  46. Bortolaia, V.; Kaas, R.S.; Ruppe, E.; Roberts, M.C.; Schwarz, S.; Cattoir, V.; Philippon, A.; Allesoe, R.L.; Rebelo, A.R.; Florensa, A.F.; et al. ResFinder 4.0 for predictions of phenotypes from genotypes. J. Antimicrob. Chemother. 2020, 75, 3491–3500. [Google Scholar] [CrossRef] [PubMed]
  47. Simner, P.J.; Miller, S.; Carroll, K.C. Understanding the Promises and Hurdles of Metagenomic Next-Generation Sequencing as a Diagnostic Tool for Infectious Diseases. Clin. Infect. Dis. 2018, 66, 778–788. [Google Scholar] [CrossRef]
  48. Shi, C.L.; Han, P.; Tang, P.J.; Chen, M.M.; Ye, Z.J.; Wu, M.Y.; Shen, J.; Wu, H.Y.; Tan, Z.Q.; Yu, X.; et al. Clinical metagenomic sequencing for diagnosis of pulmonary tuberculosis. J. Infect. 2020, 81, 567–574. [Google Scholar] [CrossRef]
  49. Zhou, X.; Wu, H.; Ruan, Q.; Jiang, N.; Chen, X.; Shen, Y.; Zhu, Y.M.; Ying, Y.; Qian, Y.Y.; Wang, X.; et al. Clinical Evaluation of Diagnosis Efficacy of Active Mycobacterium tuberculosis Complex Infection via Metagenomic Next-Generation Sequencing of Direct Clinical Samples. Front. Cell. Infect. Microbiol. 2019, 9, 351. [Google Scholar] [CrossRef] [PubMed]
  50. Fu, M.; Cao, L.J.; Xia, H.L.; Ji, Z.M.; Hu, N.N.; Leng, Z.J.; Xie, W.; Fang, Y.; Zhang, J.Q.; Xia, D.Q. The performance of detecting Mycobacterium tuberculosis complex in lung biopsy tissue by metagenomic next-generation sequencing. BMC Pulm. Med. 2022, 22, 288. [Google Scholar] [CrossRef]
  51. Liu, Y.; Wang, H.; Li, Y.; Yu, Z. Clinical application of metagenomic next-generation sequencing in tuberculosis diagnosis. Front. Cell. Infect. Microbiol. 2022, 12, 984753. [Google Scholar] [CrossRef]
  52. Kok, N.A.; Peker, N.; Schuele, L.; de Beer, J.L.; Rossen, J.W.A.; Sinha, B.; Couto, N. Host DNA depletion can increase the sensitivity of Mycobacterium spp. detection through shotgun metagenomics in sputum. Front. Microbiol. 2022, 13, 949328. [Google Scholar] [CrossRef]
  53. Kaser, M.; Ruf, M.T.; Hauser, J.; Pluschke, G. Optimized DNA preparation from mycobacteria. Cold Spring Harb. Protoc. 2010, 2010, pdb.prot5408. [Google Scholar] [CrossRef] [PubMed]
  54. Mun, B.S.; Yoon, J.; Yoon, S.Y. Optimized Method for Mycobacteria DNA Extraction from Sputum for Isothermal Amplification. Ann. Clin. Lab. Sci. 2023, 53, 476–481. [Google Scholar] [PubMed]
  55. Ratnam, S.; March, S.B. Effect of relative centrifugal force and centrifugation time on sedimentation of mycobacteria in clinical specimens. J. Clin. Microbiol. 1986, 23, 582–585. [Google Scholar] [CrossRef]
  56. Bouso, J.M.; Planet, P.J. Complete nontuberculous mycobacteria whole genomes using an optimized DNA extraction protocol for long-read sequencing. BMC Genom. 2019, 20, 793. [Google Scholar] [CrossRef]
  57. Tiew, P.Y.; Mac Aogain, M.; Ali, N.; Thng, K.X.; Goh, K.; Lau, K.J.X.; Chotirmall, S.H. The Mycobiome in Health and Disease: Emerging Concepts, Methodologies and Challenges. Mycopathologia 2020, 185, 207–231. [Google Scholar] [CrossRef]
  58. Bokulich, N.A.; Mills, D.A. Improved selection of internal transcribed spacer-specific primers enables quantitative, ultra-high-throughput profiling of fungal communities. Appl. Environ. Microbiol. 2013, 79, 2519–2526. [Google Scholar] [CrossRef]
  59. Jiang, J.; Bai, L.; Yang, W.; Peng, W.; An, J.; Wu, Y.; Pan, P.; Li, Y. Metagenomic Next-Generation Sequencing for the Diagnosis of Pneumocystis jirovecii Pneumonia in Non-HIV-Infected Patients: A Retrospective Study. Infect. Dis. Ther. 2021, 10, 1733–1745. [Google Scholar] [CrossRef] [PubMed]
  60. Limper, A.H.; Offord, K.P.; Smith, T.F.; Martin, W.J., 2nd. Pneumocystis carinii pneumonia. Differences in lung parasite number and inflammation in patients with and without AIDS. Am. Rev. Respir. Dis. 1989, 140, 1204–1209. [Google Scholar] [CrossRef]
  61. Yang, J.; Yang, F.; Ren, L.; Xiong, Z.; Wu, Z.; Dong, J.; Sun, L.; Zhang, T.; Hu, Y.; Du, J.; et al. Unbiased parallel detection of viral pathogens in clinical samples by use of a metagenomic approach. J. Clin. Microbiol. 2011, 49, 3463–3469. [Google Scholar] [CrossRef]
  62. Wylie, K.M.; Weinstock, G.M.; Storch, G.A. Emerging view of the human virome. Transl. Res. 2012, 160, 283–290. [Google Scholar] [CrossRef]
  63. Modi, A.; Vai, S.; Caramelli, D.; Lari, M. The Illumina Sequencing Protocol and the NovaSeq 6000 System. Methods Mol. Biol. 2021, 2242, 15–42. [Google Scholar] [CrossRef] [PubMed]
  64. Locher, K.; Roscoe, D.; Jassem, A.; Wong, T.; Hoang, L.M.N.; Charles, M.; Bryce, E.; Grant, J.; Stefanovic, A. FilmArray respiratory panel assay: An effective method for detecting viral and atypical bacterial pathogens in bronchoscopy specimens. Diagn. Microbiol. Infect. Dis. 2019, 95, 114880. [Google Scholar] [CrossRef] [PubMed]
  65. Cao, Z.; Sugimura, N.; Burgermeister, E.; Ebert, M.P.; Zuo, T.; Lan, P. The gut virome: A new microbiome component in health and disease. EBioMedicine 2022, 81, 104113. [Google Scholar] [CrossRef]
  66. Mirzaei, M.K.; Maurice, C.F. Menage a trois in the human gut: Interactions between host, bacteria and phages. Nat. Rev. Microbiol. 2017, 15, 397–408. [Google Scholar] [CrossRef]
  67. Moon, K.; Cho, J.C. Metaviromics coupled with phage-host identification to open the viral ‘black box’. J. Microbiol. 2021, 59, 311–323. [Google Scholar] [CrossRef]
  68. Singh, P.; Pandey, A. Community-acquired lower respiratory tract infection due to Serratia rubidaea: A rare and opportunistic pathogen. MRIMS J. Health Sci. 2025, 13, 35–38. [Google Scholar] [CrossRef]
  69. Mehdi, A.; Trifi, A.; Abbes, S.; Seghir, E.; Tlili, B.; Masseoud, L.; Noussair, A.; Ouhibi, A.; Battikh, H.; Zribi, M.; et al. Bacteremia due to Serratia rubidaea in intensive care unit: A case series. J. Med. Case Rep. 2023, 17, 482. [Google Scholar] [CrossRef]
  70. Buchan, B.W.; Windham, S.; Balada-Llasat, J.M.; Leber, A.; Harrington, A.; Relich, R.; Murphy, C.; Dien Bard, J.; Naccache, S.; Ronen, S.; et al. Practical Comparison of the BioFire FilmArray Pneumonia Panel to Routine Diagnostic Methods and Potential Impact on Antimicrobial Stewardship in Adult Hospitalized Patients with Lower Respiratory Tract Infections. J. Clin. Microbiol. 2020, 58. Available online: https://pubmed.ncbi.nlm.nih.gov/32350045/ (accessed on 6 June 2025).
  71. Ambroa, A.; Blasco, L.; Lopez-Causape, C.; Trastoy, R.; Fernandez-Garcia, L.; Bleriot, I.; Ponce-Alonso, M.; Pacios, O.; Lopez, M.; Canton, R.; et al. Temperate Bacteriophages (Prophages) in Pseudomonas aeruginosa Isolates Belonging to the International Cystic Fibrosis Clone (CC274). Front. Microbiol. 2020, 11, 556706. [Google Scholar] [CrossRef]
  72. Secor, P.R.; Burgener, E.B.; Kinnersley, M.; Jennings, L.K.; Roman-Cruz, V.; Popescu, M.; Van Belleghem, J.D.; Haddock, N.; Copeland, C.; Michaels, L.A.; et al. Pf Bacteriophage and Their Impact on Pseudomonas Virulence, Mammalian Immunity, and Chronic Infections. Front. Immunol. 2020, 11, 244. [Google Scholar] [CrossRef]
  73. Schmidt, A.K.; Fitzpatrick, A.D.; Schwartzkopf, C.M.; Faith, D.R.; Jennings, L.K.; Coluccio, A.; Hunt, D.J.; Michaels, L.A.; Hargil, A.; Chen, Q.; et al. A Filamentous Bacteriophage Protein Inhibits Type IV Pili To Prevent Superinfection of Pseudomonas aeruginosa. mBio 2022, 13, e0244121. [Google Scholar] [CrossRef]
  74. Huang, L.; Weng, B.; Gu, X.; Wang, Y.; Wang, M.; Weng, J.; Ju, Y.; Zhong, X.; Tong, X.; Li, Y. Performance of various pneumonia severity models for predicting adverse outcomes in elderly inpatients with community-acquired pneumonia. Clin. Microbiol. Infect. 2024, 30, 1426–1432. [Google Scholar] [CrossRef] [PubMed]
  75. Govender, K.N.; Street, T.L.; Sanderson, N.D.; Eyre, D.W. Metagenomic Sequencing as a Pathogen-Agnostic Clinical Diagnostic Tool for Infectious Diseases: A Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies. J. Clin. Microbiol. 2021, 59, e0291620. [Google Scholar] [CrossRef] [PubMed]
  76. Xu, Y.; Liu, D.; Han, P.; Wang, H.; Wang, S.; Gao, J.; Chen, F.; Zhou, X.; Deng, K.; Luo, J.; et al. Rapid inference of antibiotic resistance and susceptibility for Klebsiella pneumoniae by clinical shotgun metagenomic sequencing. Int. J. Antimicrob. Agents 2024, 64, 107252. [Google Scholar] [CrossRef]
  77. Hu, X.; Zhao, Y.; Han, P.; Liu, S.; Liu, W.; Mai, C.; Deng, Q.; Ren, J.; Luo, J.; Chen, F.; et al. Novel Clinical mNGS-Based Machine Learning Model for Rapid Antimicrobial Susceptibility Testing of Acinetobacter baumannii. J. Clin. Microbiol. 2023, 61, e0180522. [Google Scholar] [CrossRef]
  78. Liu, B.; Gao, J.; Liu, X.F.; Rao, G.; Luo, J.; Han, P.; Hu, W.; Zhang, Z.; Zhao, Q.; Han, L.; et al. Direct prediction of carbapenem resistance in Pseudomonas aeruginosa by whole genome sequencing and metagenomic sequencing. J. Clin. Microbiol. 2023, 61, e0061723. [Google Scholar] [CrossRef]
  79. Ruppe, E.; Cherkaoui, A.; Lazarevic, V.; Emonet, S.; Schrenzel, J. Establishing Genotype-to-Phenotype Relationships in Bacteria Causing Hospital-Acquired Pneumonia: A Prelude to the Application of Clinical Metagenomics. Antibiotics 2017, 6, 30. [Google Scholar] [CrossRef]
  80. Caroll, K.C.; Pfaller, M.A.; Landry, M.L. Manual of Clinical Microbiology; American Society for Microbiology Press: New York, NY, USA, 2019; Volume 1. [Google Scholar]
  81. Ni, J.; Yan, Q.; Yu, Y. How much metagenomic sequencing is enough to achieve a given goal? Sci. Rep. 2013, 3, 1968. [Google Scholar] [CrossRef] [PubMed]
  82. Rodriguez, R.L.; Konstantinidis, K.T. Estimating coverage in metagenomic data sets and why it matters. ISME J. 2014, 8, 2349–2351. [Google Scholar] [CrossRef]
  83. Vollmers, J.; Wiegand, S.; Kaster, A.K. Comparing and Evaluating Metagenome Assembly Tools from a Microbiologist’s Perspective—Not Only Size Matters! PLoS ONE 2017, 12, e0169662. [Google Scholar] [CrossRef]
  84. Liu, D.; Zhou, H.; Xu, T.; Yang, Q.; Mo, X.; Shi, D.; Ai, J.; Zhang, J.; Tao, Y.; Wen, D.; et al. Multicenter assessment of shotgun metagenomics for pathogen detection. EBioMedicine 2021, 74, 103649. [Google Scholar] [CrossRef]
  85. Shi, Y.; Wang, G.; Lau, H.C.; Yu, J. Metagenomic Sequencing for Microbial DNA in Human Samples: Emerging Technological Advances. Int. J. Mol. Sci. 2022, 23, 2181. [Google Scholar] [CrossRef] [PubMed]
  86. Wang, J.; Ye, J.; Yang, L.; Chen, X.; Fang, H.; Liu, Z.; Xia, G.; Zhang, Y.; Zhang, Z. Inconsistency analysis between metagenomic next-generation sequencing results of cerebrospinal fluid and clinical diagnosis with suspected central nervous system infection. BMC Infect. Dis. 2022, 22, 764. [Google Scholar] [CrossRef] [PubMed]
  87. Van Driessche, L.; Bokma, J.; Deprez, P.; Haesebrouck, F.; Boyen, F.; Pardon, B. Rapid identification of respiratory bacterial pathogens from bronchoalveolar lavage fluid in cattle by MALDI-TOF MS. Sci. Rep. 2019, 9, 18381. [Google Scholar] [CrossRef]
  88. Murphy, C.N.; Fowler, R.; Balada-Llasat, J.M.; Carroll, A.; Stone, H.; Akerele, O.; Buchan, B.; Windham, S.; Hopp, A.; Ronen, S.; et al. Multicenter Evaluation of the BioFire FilmArray Pneumonia/Pneumonia Plus Panel for Detection and Quantification of Agents of Lower Respiratory Tract Infection. J. Clin. Microbiol. 2020, 58. Available online: https://pubmed.ncbi.nlm.nih.gov/32350043/ (accessed on 6 June 2025). [CrossRef]
  89. Webber, D.M.; Wallace, M.A.; Burnham, C.A.; Anderson, N.W. Evaluation of the BioFire FilmArray Pneumonia Panel for Detection of Viral and Bacterial Pathogens in Lower Respiratory Tract Specimens in the Setting of a Tertiary Care Academic Medical Center. J. Clin. Microbiol. 2020, 58. [Google Scholar] [CrossRef]
  90. Cho, W.H.; Won, E.J.; Choi, H.J.; Kee, S.J.; Shin, J.H.; Ryang, D.W.; Suh, S.P. Comparison of AdvanSure TB/NTM PCR and COBAS TaqMan MTB PCR for Detection of Mycobacterium tuberculosis Complex in Routine Clinical Practice. Ann. Lab. Med. 2015, 35, 356–361. [Google Scholar] [CrossRef] [PubMed]
  91. Pan, X.; Yang, S.; Deighton, M.A.; Qu, Y.; Hong, L.; Su, F. A Comprehensive Evaluation of Xpert MTB/RIF Assay with Bronchoalveolar Lavage Fluid as a Single Test or Combined With Conventional Assays for Diagnosis of Pulmonary Tuberculosis in China: A Two-Center Prospective Study. Front. Microbiol. 2018, 9, 444. [Google Scholar] [CrossRef]
  92. Sow, D.; Fall, B.; Ndiaye, M.; Ba, B.S.; Sylla, K.; Tine, R.; Lo, A.C.; Abiola, A.; Wade, B.; Dieng, T.; et al. Usefulness of MALDI-TOF Mass Spectrometry for Routine Identification of Candida Species in a Resource-Poor Setting. Mycopathologia 2015, 180, 173–179. [Google Scholar] [CrossRef]
  93. Fan, L.C.; Lu, H.W.; Cheng, K.B.; Li, H.P.; Xu, J.F. Evaluation of PCR in bronchoalveolar lavage fluid for diagnosis of Pneumocystis jirovecii pneumonia: A bivariate meta-analysis and systematic review. PLoS ONE 2013, 8, e73099. [Google Scholar] [CrossRef]
  94. Imbert, S.; Meyer, I.; Palous, M.; Brossas, J.Y.; Uzunov, M.; Touafek, F.; Gay, F.; Trosini-Desert, V.; Fekkar, A. Aspergillus PCR in Bronchoalveolar Lavage Fluid for the Diagnosis and Prognosis of Aspergillosis in Patients With Hematological and Non-hematological Conditions. Front. Microbiol. 2018, 9, 1877. [Google Scholar] [CrossRef]
  95. Zeng, H.Q.; Zhang, X.B.; Cai, X.Y.; Yang, D.Y.; Lin, L.; Chen, M.J.; Guo, W.F.; Luo, X. Diagnostic value of bronchoalveolar lavage fluid cryptococcal antigen-lateral flow immunochromatographic assay for pulmonary cryptococcosis in non-HIV patients. Diagn. Microbiol. Infect. Dis. 2021, 99, 115276. [Google Scholar] [CrossRef] [PubMed]
  96. Hirsch, H.H.; Lautenschlager, I.; Pinsky, B.A.; Cardenoso, L.; Aslam, S.; Cobb, B.; Vilchez, R.A.; Valsamakis, A. An international multicenter performance analysis of cytomegalovirus load tests. Clin. Infect. Dis. 2013, 56, 367–373. [Google Scholar] [CrossRef] [PubMed]
Microorganisms 13 01338 g001
Figure 1. Flow diagram of case inclusion and exclusion. BAL: bronchoalveolar lavage. SMS: shotgun metagenomic sequencing.
Figure 1. Flow diagram of case inclusion and exclusion. BAL: bronchoalveolar lavage. SMS: shotgun metagenomic sequencing.
Microorganisms 13 01338 g001
Table 1. Comparison of microbial results from CDMs and SMS.
Table 1. Comparison of microbial results from CDMs and SMS.
NoResults of CDMResults of SMS *
Culture
(CFU/mL)		Filmarray Pneumonia Panel PCR
(Copies/mL)		Singleplex
Tests †		Taxon Above
Threshold ‡		Subdominant Taxon
1Escherichia coli (60,000)Escherichia coli (106)
Klebsiella pneumoniae (105)		MTB complexNoneStreptococcus salivarius
(111, 25.58%)
2Klebsiella pneumoniae (30,000)
Corynebacterium striatum
(≥100,000)
MTB complex		Klebsiella aerogenes (106)
Staphylococcus aureus (106)		MTB complex
CMV (128,345)
Pneumocystis jirovecii		NoneCorynebacterium striatum
(110, 26.44%)
3Candida albicans (50,000)Pseudomonas aeruginosa (106)
Escherichia coli (104)
Rhinovirus/Enterovirus		CMV (4395)Pseudomonas aeruginosa
(35, 87.5%)		Lactobacillus fermentum
(2, 5%)
4Candida tropicalis
(10,000)		Enterobacter cloacae
complex (104)		CMV (1,343,600)Candida tropicalis (5)Corynebacterium striatum
(1, 100%)
5NRFAdenovirusCMV (9665)
Pneumocystis jirovecii		NoneStaphylococcus kloosii
(2, 11.11%)
6NRFParainfluenza virusCMV (200,175)
Aspergillus(7.68)		NoneGeobacillus stearothermophilus
(1, 25%)
7Klebsiella pneumoniae (10,000)Klebsiella pneumoniae (105)NoneKlebsiella pneumoniae
(13, 30.23%)		Serratia marcescens
(3, 6.98%)
8NRFKlebsiella pneumoniae (105)CMV (1015)NoneKlebsiella pneumoniae
(8, 17.78%)
9NRFKlebsiella pneumoniae (105)NoneNoneNone
10NRFHaemophilus influenzae (106)
Metapneumovirus		NoneHaemophilus influenzae
(90, 59.60%) 		Haemophilus parasuis
(2, 1.33%)
11NRFHaemophilus influenzae (105)
Influenza A		CMV (<257)Haemophilus influenzae
(5, 31.25%) 		Pasteurella multocida
(2, 12.5%)
12Pseudomonas aeruginosa
(≥100,000)		Pseudomonas aeruginosa (106) CMV (<257)
Pneumocystis jirovecii
Aspergillus (1.08)		Pseudomonas aeruginosa
(59, 75.64%) 		Pseudomonas sp.
(6, 7.69%)
13Acinetobacter baumannii
(≥100,000)		Acinetobacter calcoaceticus-
baumannii complex (106)		Aspergillus (2.27)Acinetobacter baumannii
(141, 66.82%)		Acinetobacter nosocomialis
(11, 5.21%)
Candida albicans (6)
14Pseudomonas aeruginosa
(≥100,000)		Acinetobacter calcoaceticus-
baumannii complex (≥107)
Pseudomonas aeruginosa (≥107)
Escherichia coli (106)
Serratia marcescens (106)
Klebsiella pneumoniae 105)		CMV (<257)Pseudomonas aeruginosa
(23,869, 86.63%) 		Acinetobacter baumannii
(347, 1.26%)
15Pseudomonas aeruginosa
(≥100,000)		Pseudomonas aeruginosa (106) CMV (2065)
Aspergillus (1.90)		Pseudomonas aeruginosa
(24, 75%) 		Pseudomonas sp.
(2, 6.25%)
16Stenotrophomonas maltophilia
(≥100,000)		Parainfluenza virusCMV (10,780)Stenotrophomonas maltophilia
(147, 44.41%)		Stenotrophomonas pavanii
(26, 7.85%)
Detected microorganisms that were treatment targets are underlined. For pathogens detected by both culture and the PCR, only the culture results are underlined. * For SMS results, the read numbers and their fractions among 16S reads were recorded in parentheses for bacteria, while only the read numbers were recorded in parentheses for fungi. † Singleplex tests included PCR for MTB, CMV, and Pneumocystis jirovecii as well as an antigen test for Aspergillus. Quantitative values for CMV PCR were recorded in parentheses as copies/mL, and antigen test results for Aspergillus were also noted in parentheses. ‡ For SMS interpretation, a threshold of 30% relative abundance, based on the fraction of 16S reads, was used as per previous studies. Abbreviations: no., number; CDM, conventional diagnostic method; SMS, shotgun metagenomic sequencing; MTB; Mycobacterium tuberculosis; CMV, cytomegalovirus; NRF, normal respiratory flora.
Table 2. Antibiotic resistance genes detected in case 2 with Klebsiella pneumoniae.
Table 2. Antibiotic resistance genes detected in case 2 with Klebsiella pneumoniae.
Antibiotic GroupResistance GeneGene Detection StatusAntibiotic SusceptibilityConsistency *
CarbapenemNoneNot detectedDoripenem (S), Ertapenem (S), Imipenem (S), Meropenem (S)Consistent
AminoglycosideAAC(6′)-IaDetectedAmikacin (S), Gentamicin (S)Inconsistent
SulfonamideSul1DetectedTrimethoprim/Sulfamethoxazole (R)Consistent
* The antibiotic resistance gene aligned with the resistance profile of the cultured isolates. Abbreviations: S, susceptible; R, resistant.
Table 3. Antibiotic resistance genes detected in case 14 with Pseudomonas aeruginosa.
Table 3. Antibiotic resistance genes detected in case 14 with Pseudomonas aeruginosa.
Antibiotic GroupResistance GeneGene Detection
Status		Antibiotic SusceptibilityConsistency *
CarbapenemAXC-1, OXA-217DetectedDoripenem (S), Imipenem (S), Meropenem (S)Inconsistent
AminoglycosideAAC(6′)-Ia, cpxA, smeRDetectedAmikacin (S), Tobramycin (I)Inconsistent
Fluoroquinolone and TetracyclineMexI, H-NSDetectedLevofloxacin (R),
Tetracycline (R)		Consistent
Penicillin derivatives and Cephalosporins OXA-217, smeR, H-NS, CTX-M-15DetectedAmpicillin (R), Piperacillin (R),
Cefotaxime (R), Cefepime (S), Ceftazidime (S)		Inconsistent
* The antibiotic resistance gene aligned with the resistance profile of the cultured isolates. Abbreviations: S, susceptible; R, resistant; I, indeterminate.
Table 4. Thresholds for identifying clinically significant microbe using SMS.
Table 4. Thresholds for identifying clinically significant microbe using SMS.
DiseaseSequencingPrinciples of Setting
Thresholds		Thresholds for Identifying Pathogen by SMSStudy
BacteriaMycobacteriaFungiViruses
Suspected pulmonary infection BGISEQ
(BGI, Shenzhen, China) 		≥50 unique reads from a single species ≥1 unique read from MTBC ≥50 unique reads from a single species ≥50 unique reads from a single species [19]
Suspected pulmonary infection BGISEQ-100 (BGI, China) Reads number ≥ 50 and pathogen detected by traditional method>30% relative abundance at the genus level ≥1 unique read from MTB SMRN ≥ 3[18]
Suspected pulmonary infection MiniSeq (Illumina, USA) >30% relative abundance at the genus level, or
histopathological examination and/or culture positive with ≥50 unique reads of a single species 		≥1 unique read from MTB >30% relative abundance at the genus level, or histopathological examination and/or culture positive with ≥50 unique reads of a single species >30% relative abundance at the genus level [17]
Suspected pneumonia (immunocompromised) Agilent 2100 Bioanalyzer (Thermo Fisher Scientific, USA)Regardless of coverage rate, oral commensals were not defined as CSMs unless they were deemed to be significant by the physicians or proven otherwise Coverage rate ≥ 10 times any other microbes ≥1 unique read from MTB
Mapping read number in the top 10 in the bacteria list of NTM 		Coverage rate ≥ 5 times any other fungus Coverage rate ≥ 10 times any other microbes [23]
Pneumonia Bioelectron Seq 4000 (CapitalBio Corporation, Beijing, China)Clinically pathologic microorganism is defined;
-Definite: SMS result is consistent with results from CDMs (culture, nucleic acid-based testing, and pathological examination)  
Probable: SMS pathogen is likely the cause of pneumonia according to clinical, radiologic, or laboratory findings, but the SMS result was consistent with CDMs.		Coverage rate ≥ 10 times any other microbes ≥1 unique read from MTB Coverage rate ≥ 5 times any other fungus Coverage rate of species level ≥ 10 times any other microbes [22]
Severe pneumonia (immunocompromised) KAPA Library Quantification 75-cycle sequencing kit (Illumina, USA) >30% Relative abundance at the genus level,
Reads number ≥ 50 from a single species and pathogen detected by culture 		[20]
Community-acquired pneumonia Nextseq 550Dx (Illumina, USA) RPM counts ≥ 5 times the values of the NEC Coverage of ≥3 non-overlapping regions on the genome [25]
Community-acquired pneumonia NextSeq CN500 (Illumina, USA) Reads number ≥ 50 or pathogen detected by culture ≥1 unique read from MTB
Mapping read number in the top 10 in the bacteria list of NTM 		≥3 reads mapped to pathogen species, or supported by clinical culture ≥3 reads mapped to pathogen species, or supported by clinical culture [24]
Severe community-acquired pneumonia (immunocompromised) N/AReads number ≥ 50 from a single species and pathogen detected by culture >30% Relative abundance at the genus level,
Coverage rate ≥ 10 times any other bacteria 		≥1 unique read from MTB>30% relative abundance at the genus level;
Coverage rate ≥ 5 times any other fungus 		[26]
Abbreviations: SMS, shotgun metagenomic sequencing; RPM, read per million; NEC, negative extraction control; MTBC, Mycobacterium tuberculosis complex; MTB, Mycobacterium tuberculosis; SMRN, stringently mapped read number; CSMs, clinically significant microbes; NTM, non-tuberculous mycobacteria; CDM, conventional diagnostic method; N/A, not accessible.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Cho, H.-e.; Kim, M.J.; Choi, J.; Sohn, Y.-H.; Lee, J.J.; Park, K.S.; Cho, S.Y.; Park, K.-H.; Kim, Y.J. Shotgun Metagenomic Sequencing Analysis as a Diagnostic Strategy for Patients with Lower Respiratory Tract Infections. Microorganisms 2025, 13, 1338. https://doi.org/10.3390/microorganisms13061338

AMA Style

Cho H-e, Kim MJ, Choi J, Sohn Y-H, Lee JJ, Park KS, Cho SY, Park K-H, Kim YJ. Shotgun Metagenomic Sequencing Analysis as a Diagnostic Strategy for Patients with Lower Respiratory Tract Infections. Microorganisms. 2025; 13(6):1338. https://doi.org/10.3390/microorganisms13061338

Chicago/Turabian Style

Cho, Ha-eun, Min Jin Kim, Jongmun Choi, Yong-Hak Sohn, Jae Joon Lee, Kyung Sun Park, Sun Young Cho, Ki-Ho Park, and Young Jin Kim. 2025. "Shotgun Metagenomic Sequencing Analysis as a Diagnostic Strategy for Patients with Lower Respiratory Tract Infections" Microorganisms 13, no. 6: 1338. https://doi.org/10.3390/microorganisms13061338

APA Style

Cho, H.-e., Kim, M. J., Choi, J., Sohn, Y.-H., Lee, J. J., Park, K. S., Cho, S. Y., Park, K.-H., & Kim, Y. J. (2025). Shotgun Metagenomic Sequencing Analysis as a Diagnostic Strategy for Patients with Lower Respiratory Tract Infections. Microorganisms, 13(6), 1338. https://doi.org/10.3390/microorganisms13061338

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop