Abstract:
Metagenomics has expanded our ability to study complex microbial communities, yet its
practical application faces barriers including the requirement of programming expertise,
and complex command-line workflows that hinder rapid analysis and interpretation.
These challenges can affect collaborative research between wet-lab scientists and
bioinformaticians, often creating bottlenecks in time-sensitive studies.
This dissertation addresses these challenges by describing accessible computational
tools specifically optimized for long-read sequencing data and real-time monitoring
of microbial communities. The research includes an investigation of chain-elongating
microbiomes in bioreactors, revealing key microbial players and metabolic pathways
involved in medium-chain carboxylate production. Through metagenomic analyses,
the collaborative work identifies the role of oxygen in n-caprylate production and the
complex interactions between aerobic and anaerobic species.
The dissertation introduces MMonitor, a novel software platform that combines compu-
tational pipelines with interactive visualization for real-time analysis of metagenomic
data from Oxford Nanopore Technologies sequencing. MMonitor’s capabilities are
demonstrated through applications in tracking bioreactor microbiomes, as well as in
contamination control for artificial intelligence training data. Additional methodological
contributions include QuickBinDM, which accelerates long-read metagenomic binning
through pre-screening approaches, and GeneGone, a web application for the validation
of gene deletion experiments.
Together, these tools and analyses advance our understanding of complex microbial
communities while making metagenomic analysis more accessible to researchers. The
work has implications for bioinformaticians, biologists and researchers from related
fields interested in metagenomics and the tracking of microbial communities.