Microbiome · 2018 · 123 citations · 59 references
Life SciencesEcologyMicrobiologyDaylightingMicrobial EcologyMicrobiome ResearchBacterial EcologyDaylight ModelingEnvironmental MicrobiomesMicrobial Community DynamicsMicrobiome Data AnalysisMicrobiome DynamicsAirborne Microbial Community StructureDominant SpeciesEcological Impacts Of LightHousehold Dust ExposureIndoor Air MicrobiologyLight ExposureUltraviolet Light EffectsUltraviolet Light Exposure
BACKGROUND: Microbial communities associated with indoor dust abound in the built environment. The transmission of sunlight through windows is a key building design consideration, but the effects of light exposure on dust communities remain unclear. We report results of an experiment and computational models designed to assess the effects of light exposure and wavelengths on the structure of the dust microbiome. Specifically, we placed household dust in replicate model "rooms" with windows that transmitted visible, ultraviolet, or no light and measured taxonomic compositions, absolute abundances, and viabilities of the resulting bacterial communities. RESULTS: Light exposure per se led to lower abundances of viable bacteria and communities that were compositionally distinct from dark rooms, suggesting preferential inactivation of some microbes over others under daylighting conditions. Differences between communities experiencing visible and ultraviolet light wavelengths were relatively minor, manifesting primarily in abundances of dead human-derived taxa. Daylighting was associated with the loss of a few numerically dominant groups of related microorganisms and apparent increases in the abundances of some rare groups, suggesting that a small number of microorganisms may have exhibited modest population growth under lighting conditions. Although biological processes like population growth on dust could have generated these patterns, we also present an alternate statistical explanation using sampling models from ecology; simulations indicate that artefactual, apparent increases in the abundances of very rare taxa may be a null expectation following the selective inactivation of dominant microorganisms in a community. CONCLUSIONS: Our experimental and simulation-based results indicate that dust contains living bacterial taxa that can be inactivated following changes in local abiotic conditions and suggest that the bactericidal potential of ordinary window-filtered sunlight may be similar to ultraviolet wavelengths across dosages that are relevant to real buildings.
59
R: A Language and Environment for Statistical Computing
R Core Team · 2000 · 353K citations · Full text
Natural Sciences, Computer Science, Statistical Computing +10
DADA2: High-resolution sample inference from Illumina amplicon data
Benjamin J. Callahan, Paul J. McMurdie, Michael Rosen et al. · Nature Methods · 2016 · 36.3K citations
Laurens van der Maaten, Geoffrey E. Hinton · Journal of Machine Learning Research · 2008 · 35.7K citations