Informed interpretation of metagenomic data by StrainPhlAn enables strain retention analyses of the upper airway microbiome

Abstract

The usage of 16S rRNA gene sequencing has become the state-of-the-art method for the characterization of the microbiota in health and respiratory disease. The method is reliable for low biomass samples due to prior amplification of the 16S rRNA gene but has limitations as species and certainly strain identification is not possible. However, the usage of metagenomic tools for the analyses of microbiome data from low biomass samples is not straight forward, and careful optimization is needed. In this work, we show that by validating StrainPhlAn 3 results with the data from bacterial cultures, the strain-level tracking of the respiratory microbiome is feasible despite the high content of host DNA being present when parameters are carefully optimized to fit low biomass microbiomes. This work further proposes that strain retention analyses are feasible, at least for more abundant species. This will help to better understand the longitudinal dynamics of the upper respiratory microbiome during health and disease.

Keywords: bacterial culture; genome analysis; metagenomics; respiratory tract; strain resolution.

Fig 1

Benchmarking the different set of parameters. The default set of parameters is illustrated in run No. 0 and is BREADTH_THRESHOLD 80, TRIM_SEQUENCES 50, MARKER_IN_N_SAMPLES 80, and SAMPLE_WITH_N_MARKERS 20. Panels a, b, c, and d represent benchmarking for NP data; e and f for OP data. The default (No. 0) and 17 different sets of parameters (No. 1–17) are illustrated on the x-axis, and their characteristics are described in Table S1. Gray = specificity, yellow = sensitivity, and green = F1-score. The optimized sets of parameters are shown at No. 7 (indicated in red).

Fig 2

StrainPhlAn-derived trees from NP data sets. Trees are StrainPhAn 3 generated for (a) S. pneumoniae, (b) M. catarrhalis, (c) S. aureus. and (d) H. influenzae from NP data and were using the optimized parameters of StrainPhlAn 3. Samples are indicated in different colors depending on the presence/absence of culture data [i.e. purple = found in StrainPhlAn but not culture, yellow = found with StrainPhlAn and culture, and gray = found with StrainPhlAn but no culture data available (NA)].

Fig 3

Phylogenetic trees of S. aureus from OP data sets. On the left, core genome tree from WGS data of S. aureus isolates. On the right, the phylogenetic tree was created from metagenomic data of OP samples using the adjusted parameters of StraiPhlAn. Sample IDs are indicated in black.

Fig 4

Strain retention analyses based on normalized genetic distances of four different species of the NP data set. The all-versus-all normalized genetic distances have been separately calculated for S. pneumoniae, M. catarrhalis, H. influenzae, and S. aureus. Values were binned in intervals of 0.1. Bins with values 0.0–0.1 (for S. aureus and H. influenzae) and 0.0–0.2 (for S. pneumoniae and M. catarrhalis) were defined for strain retention (dotted lines; see text for details).

Fig 5

Clustering and longitudinal visualization of S. aureus strains of OP samples. Phylogenetic tree was created with StrainPhlAn using the new parameters from metagenomic data, highlighting 10 clusters (1-J) (Fig. 5A). Strain retention of S. aureus was investigated in a total of fifteen individuals with CF, sampled for up to the first 10 years of life (Fig. 5B). Strain retainment was investigated by (i) extracting the MLST from the WGS from S. aureus (SA) and (ii) using the clustering information from metagenomic data (shown as SA_clusters from A-J with different colors). For the detection of SA we used four categories. No S. aureus found by culture and metagenomic sequencing (indicated by an empty circle), S. aureus found by culture and metagenomic sequencing (filled triangle), S. aureus found by culture but not metagenomic sequencing (empty triangle), and S. aureus found by metagenomic sequencing but not by culture (filled circle). StrainPhlAn clusters are indicated with different colors. Numbers reflect the sequence types from MLST found in the respective S. aureus isolates. As for triangles without numbers, S. aureus by culture has been reported, but the isolate has not been kept for WGS.