Back to search

University of Minnesota

Elucidating Fungal Traits with the Help of Field Collection Data and Novel Data Science Tools

Abstract

dc:description.abstract

Fungi are critical to the functioning of most ecosystems and a better understanding of their traits, like habitat, host, or substrate associations, could help inform land management decisions, improve models for biogeochemical cycles, or even help select better strains for industrial processes. Efforts to study fungal traits, however, often require considerable time and resources to conduct field and laboratory work, which ultimately restricts the spatial and temporal comprehensiveness of any study or might encourage focus on a particular taxonomic group. This means that existing conclusions regarding fungal traits are typically based on limited data with spatial, temporal, or even taxonomic biases. One solution is to utilize existing digitized field collection and observation data (e.g., taxon, location, date/time, habitat, substrate) that is aggregated and made available through online services like The Global Biodiversity Information Facility and The Mycology Collections Portal. These data are comprised of millions of globally distributed and temporally comprehensive (i.e., ranging over one hundred years) occurrence records based on digitized fungarium specimens and community-science observations (e.g., iNaturalist, Mushroom Observer). Importantly, some metadata (i.e., habitat or substrate notes) included with these records can be used to determine if a specimen exhibited a particular trait. Despite its potential, these data have challenges that restrict its use in comprehensive trait analyses. These challenges include outdated taxonomy, misspellings of taxon and location names, invalid dates or GPS coordinates, heterogeneous data format, and a general lack of tools for identifying traits in record metadata and analyzing trait patterns. This dissertation introduces and demonstrates novel tools for overcoming the challenges of field collection data and enabling their use in comprehensive fungal trait studies. It does so by first highlighting the challenges of examining fungal traits (e.g., fire association) in the laboratory and how field collection data can be used to enhance experimental design. Specifically, a set of quantitative decay experiments were conducted to evaluate the theory that some fungi are strongly associated with fire-affected environments because they have an enhanced ability to degrade charred plant residues. This fire-adaptive mechanism could have serious implications for carbon cycling, as the natural degradation of charred plant biomass is not well understood, and this material is often described as a carbon sink. Neurospora crassa was chosen for testing due to its well-documented association with charred plant biomass in burned environments and the fact that it is a model fungus with well-developed genetic tools that may enable more detailed studies of how it degrades charred plant material. However, N. crassa was shown to have no enhanced ability to degrade charred plant material compared to the non-fire-associated species tested. The next logical step towards validating this fire-adaptive mechanism would be to test more fire-associated species; however, many species have documented fire associations and the number of species that can be tested in a single laboratory experiment is restricted by time and resources. Additionally, many studies documenting fire association (i.e., relative abundances of species in burned vs non burned environments) are not spatially, temporally, or taxonomically comprehensive and there are often conflicting results between studies. All these issues pose a challenge to picking species with the strongest association with fire-affected environments and the highest likelihood of having an enhanced ability to degrade charred plant residues. To identify strongly fire-associated fungal species in a more comprehensive and quantitative way, field collection data from all across the United States (1 859 456 total records) were used to quantify and compare fire association (i.e., fire-associated records proportion (per species): fire-associated records/(fire- + non-fire-associated records)) among Agaricomycotina and Pezizomycotina fungi. Because of the issues associated with field collection data (outlined above), custom R scripts were created to overcome these issues and enable the use of field collection records for analyzing fire association. Using these methods, species were successfully ranked, and strongly fire-associated species were identified. Because of the success of this analysis and the apparent potential of using this methodology for studying fungal traits, in general, the custom R scripts were incorporated into a new R package called FUNGARIUM. The motivation here was to make these tools available to everyone and inspire similarly comprehensive analyses of other fungal traits using this methodology. Based on the success of using field collection data and the novel R package FUNGARIUM to investigate fungal fire association, this same approach was used to investigate another important fungal trait (i.e., habitat and substrate affinities of brown and white rot wood decay fungi). Brown rot (BR) and white rot (WR) wood decay fungi differ in how they degrade wood, with BR fungi generally avoiding wood lignin, unlike WR. This has carbon cycling implications, as BR decay may equate to more carbon deposition, relative to WR, due to the deposition of undecayed lignin residues. Despite this significance, comprehensive analyses of the relative distributions of BR and WR fungi are lacking. Some evidence suggests that BR fungi have a higher relative abundance in coniferous habitats and that this is due to a preference for conifer substrates; however, this has not been validated at a globally comprehensive scale. Using globally comprehensive field collection data and the R package FUNGARIUM, a global correlation (R² = 0.45) between brown rot fungi proportion (i.e., brown rot records/(brown + white rot records)) and conifer tree proportion (i.e., conifer records/(conifer + angiosperm records)) was identified. Additionally, BR fungi were found to have a stronger affinity for conifer substrates relative to WR (i.e., 61% of BR and 19% of WR species were identified as conifer substrate specialists). Overall, this study represents the most comprehensive analysis of habitat and substrate affinities among BR and WR fungi to date, and, again, demonstrates the great utility of field collection data the novel R package FUNGARIUM. Hopefully, this work, and the other studies presented in this dissertation, will inspire and facilitate future comprehensive studies of fungal traits.

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Simpson, Hunter

Subjects

dc:subject × 6

Rights

Language dc:language.iso
en

Identifiers

dc:identifier.*
Dc Identifier Citation
Simpson, Hunter. (2024). Elucidating Fungal Traits with the Help of Field Collection Data and Novel Data Science Tools. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/283151.

Chain of custody

source
Harvested from
University of Minnesota
Base URL
conservancy.umn.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Simpson, Hunter. Elucidating Fungal Traits with the Help of Field Collection Data and Novel Data Science Tools. 2024. https://hdl.handle.net/11299/283151