Decoding Microbial Dynamics: Computational Pipelines for Multi-omics Ecosystem Analysis

Abstract

Microbial communities play a fundamental role in the regulation of ecosystem processes, yet understanding how they respond to environmental fluctuations remains a key challenge in microbial ecology. Thanks to the advances in high-throughput sequencing and mass spectrometry technologies, the generation and application of multi-omics approaches for the study of biological systems has dramatically increased in recent years leading to unprecedented insights into microbial dynamics. However, the complexity and scale of these datasets has resulted in the need for robust computational frameworks to extract meaningful biological information. This dissertation sits at the intersection of computational biology and microbial ecology by developing and applying novel data analysis pipelines to investigate microbial community structure, function, and resilience. Chapter 1 introduces MetaboDirect and MetaboTandem, two new data analysis pipelines designed to streamline the analysis of high-resolution metabolomics datasets. These tools address critical barriers to the accessibility of metabolomics, and provide means to overcome challenges with data preprocessing and annotation. Chapters 2 and 3 apply these and other computational tools to study microbial dynamics in two distinct ecosystems, peatlands and arid soils. In the peatlands, tracking carbon from litter-derived metabolites using stable isotope-assisted metabolomics revealed that litter inputs result only in a small, transient priming effect suggesting that palsa peatland carbon storage may remain stable under future climate scenarios. In arid soils, the integration of metagenomics, metatranscriptomics, and metabolomics identified metabolic network reorganization, community assembly processes, and conserved genomic traits as key components of microbial stability under extreme environmental fluctuations. Together, these studies demonstrate how the development and application of computational tools can advance ecological understanding of microbial ecosystems. The findings have broad implications for predicting ecosystem responses to climate change and for providing data analysis frameworks applicable across different ecosystems and environmental conditions.