The house dust microbiota in the Drakenstein Child Health Study

• Main Author: Duyver, Menna
• Other Authors: Nicol, Mark
• Format: Dissertation
• Language: English
• Published: University of Cape Town 2015
• Subjects: Medical Microbiology
• Online Access: http://hdl.handle.net/11427/15516

Introduction: The indoor home environment comprises many niches that are occupied by bacterial communities. The composition of these bacterial communities may be influenced by numerous factors such as number of occupants, pets, season and location. Understanding the house dust microbial community is vital to understanding its' influence on human respiratory health. Aims: The aims of the studies described in this MSc dissertation were to: 1) evaluate the performance of ten commercial nucleic acid extraction kits on dust samples; 2) optimise dust removal from electrostatic dustfall collectors (EDC); 3) determine the bacterial composition of house dust using 16S rRNA gene sequencing and 4) determine those factors influencing the bacterial composition of house dust by performing bioinformatic and data analysis on the sequenced dust samples. Methods: In order to study the microbial content of house dust, an efficient DNA extraction protocol was required. Ten commercial nucleic acid purification protocols were evaluated on their ability to efficiently extract good quality DNA from very low quantities (20 mg) of wet bulk house dust. For the purpose of this study, EDCs were used to collect settled dust from homes of participants in the Drakenstein Child Health Study (DCHS). Electrostatic Dustfall Collectors were placed twice within the same household, approximately 6 months apart, spanning two seasons. The Z/R Fungal/Bacterial DNA MicroprepTM (ZMC) protocol was used to extract DNA from dust removed from EDCs. The V4 region of the 16S rRNA gene was amplified and sequenced using the Illumina MiSeq platform to determine the bacterial taxonomic composition of the house dust samples. A custom python wrapper that meshes a set of tools integrated into a computationally efficient workflow, known as the YAP pipeline was used to classify 16S rRNA sequences into bacterial taxonomies. Based on 97% sequence similarity, the pre-processed sequences were assigned to Operational Taxonomic Units (OTU). R software together with RStudio software was used for all statistical analysis and graphical representations of the data.