Metadata and Phyloseq R objects from 16S and ITS1 for paper: Residential Airborne Microbiota and Childhood LRTI

These ready to load phyloseq R S4 objects contain the ASV table, taxonomy table and sample metadata (bacterial 16S V5-V6 and fungi ITS1).  This dataset was build using the DaDa2 (version 1.22.0) and phyloseq (version 1.38.0) R packages using SILVA (v128) and UNITE (v7.2) microbial databases. The accompanying (peer-reviewed) scientific article / sequencing data can be found here:  doi.org   Study abstract: Adekunle G. FAKUNLE, Nkosana JAFTA, Alex BOSSERS, Inge M. WOUTERS, Warner van KERSEN, Rajen N. NAIDOO, Lidwien A.M SMIT Residential microbial composition likely contributes to the development of lower respiratory tract infections (LRTI) among children, but the association is poorly understood. We aimed to study the relationship between the indoor airborne dust bacterial and fungal microbiota and childhood LRTI in Ibadan, Nigeria. Ninety-eight children under the age of five years hospitalized with LRTI were recruited and matched by age (±3 months), sex, and geographical location to 99 community-based controls without LRTI. Participants' homes were visited and sampled over a 14-day period for airborne house dust using electrostatic dustfall collectors (EDC). In airborne dust samples, the composition of bacterial and fungal communities was characterized by a meta-barcoding approach using amplicons targeting simultaneously the bacterial 16S rRNA gene and the internal-transcribed-spacer (ITS) region-1 of fungi in association with the SILVA and UNITE database respectively. A 100-unit change in house dust bacterial, but not fungal, richness (OR 1.06; 95%CI 1.03-1.10) and a 1-unit change in Shannon diversity (OR 1.92; 95%CI 1.28-3.01) were both independently associated with childhood LRTI after adjusting for other indoor environmental risk factors. Beta-diversity analysis showed that bacterial (PERMANOVA p < 0.001, R2 = 0.036) and fungal (PERMANOVA p < 0.001, R2 = 0.028) community composition differed significantly between homes of cases and controls. Pair-wise differential abundance analysis using both DESEq2 and MaAsLin2 consistently identified the bacterial phyla Deinococcota (Benjamini-Hochberg (BH) adjusted p-value <0.001) and Bacteriodota (BH-adjusted p-value = 0.004) to be negatively associated with LRTI. Within the fungal microbiota, phylum Ascomycota abundance (BH adjusted p-value <0.001) was observed to be directly associated with LRTI, while Basidiomycota abundance (BH adjusted p-value <0.001) was negatively associated with LRTI. Our study suggests that early-life exposure to certain airborne bacterial and fungal communities is associated with LRTI among children under the age of five years.