School of Civil and Environmental Engineering
Ph.D. Thesis Defense Announcement
DEVELOPING META’OMIC METHODS FOR FORENSIC ANALYSIS OF WASTEWATER INFRASTRUCTURE PERFORMANCE
By Blake Greer Lindner
Advisor:
Dr. Kostas Konstantinidis (CEE)
Committee Members:
Dr. Ching-Hua Huang (CEE) | Dr. Katherine Graham (CEE)
Dr. Ameet Pinto (CEE) | Dr. Blythe Layton (Clean Water Services)
Date and Time: June 6, 2024. 1PM EDT
Location: ES&T L1116 | Online at https://bit.ly/4bogo5T
Although the application of metagenomic and metatranscriptomic (i.e., meta'omic) methods to sanitation issues is increasing rapidly, these methods remain inadequately standardized and in need of evaluation with more convincing benchmarks to see widespread adoption in engineering practice. There are currently no reproducible approaches to establish detection limits or estimate microbial population proportions that marry theory and in silico experiments with empirical data. To address these limitations, this thesis begins with the construction of a theoretical framework for estimating meta’omic limits of detection and population proportioning. Building on this theory, the remaining work advances new methods for two forensic uses within the context of wastewater infrastructure:First (1), a metagenomic approach for fecal source tracking (FST) was developed using data produced from laboratory scale mesocosms simulating fecal contamination events. These results were used to construct an open source bioinformatic pipeline called, “SourceApp”. SourceApp was found to readily distinguish between sewage and septage contamination, among several other common fecal sources, both in terms of presence/absence and portions. The second (2) use case this thesis investigates is root cause analysis of process upsets during enhanced biological phosphorus removal (EBPR) operations using metagenomic and metatranscriptomic data. Statistical models to predict process performance were constructed using biological features returned by the meta’omic data and interrogated for feature importance. Additionally, differential expression analysis data revealed gene-level dynamics occurring during process upsets. Combined, these efforts reveal microbial populations and specific genes which were sensitive to process conditions linked to observed upsets and may serve as sensitive biomarkers for routine performance monitoring. Overall, these efforts elucidate promising methodologies for better integrating meta’omic data into important investigative aims within environmental engineering such as FST and EBPR biomonitoring.