Discovery of Cell Wall and Cuticle Gene Candidates for Biofuel Feedstock Improvement in Populus trichocarpa (Black Cottonwood) Using Systems Biology Approaches

Climate change threatens agriculture and bioenergy production, necessitating the development of resilient, sustainable feedstocks that can thrive on marginal lands with limited resources. Understanding the genetic architecture underlying complex traits such as biomass accumulation and stress tolerance is essential for crop improvement. Populus trichocarpa (black cottonwood) is an important biofuel feedstock and model woody plant with extensive genomic resources and natural populations spanning diverse environmental gradients. This dissertation uses systems and network biology methods to characterize genomic variation in P. trichocarpa and link genetic loci with potentially impacted traits, furthering our understanding of the contribution of natural variation to environmental adaptation and identifying targets for feedstock improvement.

Chapter 2 leverages a genome-wide association study (GWAS) to identify single nucleotide polymorphisms (SNPs) associated with cell wall metabolite concentrations. Then, network methods are used to identify novel candidate genes involved in regulation of cell wall biosynthesis – a key trait for plant growth, stress response, and biofuel conversion efficiency. High-priority candidates have since been experimentally validated, demonstrating the power of network-based approaches for gene discovery.

In Chapter 3, I use two detection algorithms to call 794,619 and 877,216 large SVs in 1,484 wild P. trichocarpa accessions. Population structure analyses revealed distinct subgroups reflecting post-glacial demographic history, with northern populations showing reduced diversity consistent with rapid expansion from refugia. This SV catalog provides a resource for further studies on the impacts of large-scale genomic rearrangements.

Chapter 4 investigates associations between SVs and climate variables such as water deficit, soil moisture, and temperature. Using relaxed GWAS thresholds combined with network filtering strategies, this approach identified 13,124 and 11,297 SV-climate associations, respectively, with different subtypes capturing distinct evolutionary signals. Network methods GRIN and MENTOR retained 5,471 biologically relevant genes, three of which (PtKCS) were identified as high-priority candidates. Located within a complex SV region, PtKCS genes encode enzymes essential for cuticular wax biosynthesis and contribute to population-level variation in drought and pathogen response.

Collectively, this work demonstrates how integrating population genomics, association mapping, and network biology can identify genetic candidates relevant for environmental adaptation in forest trees and improvement of feedstock sustainability.

Last upload had an error in the TOC.