Deep Learning For Supervised Classification Of Temporal Data In Ecology, César Capinha, Ana Ceia-Hasse, Andrew M. Kramer, Christiaan Meijer

Integrative Biology Faculty and Staff Publications

Temporal data is ubiquitous in ecology and ecologists often face the challenge of accurately differentiating these data into predefined classes, such as biological entities or ecological states. The usual approach consists of transforming the time series into user-defined features and then using these features as predictors in conventional statistical or machine learning models. Here we suggest the use of deep learning models as an alternative to this approach. Recent deep learning techniques can perform the classification directly from the time series, eliminating subjective and resource-consuming data transformation steps, and potentially improving classification results. We describe some of the deep learning …

Global Economic Costs Of Aquatic Invasive Alien Species, Ross N. Cuthbert, Zarah Pattison, Nigel G. Taylor, Laura Verbrugge, Christophe Diagne, Danish A. Ahmed, Boris Leroy, Elena Angulo, Elizabeta Briski, César Capinha, Jane A. Catford, Tatenda Dalu, Franz Essl, Rodolphe E. Gozlan, Phillip J. Haubrock, Melina Kourantidou, Andrew M. Kramer, David Renault, Ryan J. Wasserman, Franck Courchamp

Integrative Biology Faculty and Staff Publications

Much research effort has been invested in understanding ecological impacts of invasive alien species (IAS) across ecosystems and taxonomic groups, but empirical studies about economic effects lack synthesis. Using a comprehensive global database, we determine patterns and trends in economic costs of aquatic IAS by examining: (i) the distribution of these costs across taxa, geographic regions and cost types; (ii) the temporal dynamics of global costs; and (iii) knowledge gaps, especially compared to terrestrial IAS. Based on the costs recorded from the existing literature, the global cost of aquatic IAS conservatively summed to US$345 …

Deep Learning For Supervised Classification Of Temporal Data In Ecology, César Capinha, Ana Ceia-Hasse, Andrew M. Kramer, Christiaan Meijer

Integrative Biology Faculty and Staff Publications

Temporal data is ubiquitous in ecology and ecologists often face the challenge of accurately differentiating these data into predefined classes, such as biological entities or ecological states. The usual approach consists of transforming the time series into user-defined features and then using these features as predictors in conventional statistical or machine learning models. Here we suggest the use of deep learning models as an alternative to this approach. Recent deep learning techniques can perform the classification directly from the time series, eliminating subjective and resource-consuming data transformation steps, and potentially improving classification results. We describe some of the deep learning …

Spatial Spread Of White-Nose Syndrome In North America, 2006-2018, Andrew M. Kramer, Alex Mercier, Sean Maher, Yaw Kumi-Ansu, Sarah Bowden

Integrative Biology Faculty and Staff Publications

White-nose syndrome has caused massive mortality in multiple bat species and spread across much of North America, making it one of the most destructive wildlife diseases on record. This has also resulted in it being one of the most well-documented wildlife disease outbreaks, making it possible to look for changes in the pattern of spatial spread over time. We fit a series of spatial interaction models to the United States county-level observations of the pathogenic fungus, Pseudogymnoascus destructans, that causes white-nose syndrome. Models included the distance between caves, cave abundance, measures of winter length and winter onset, and species …

Deep Learning For Supervised Classification Of Temporal Data In Ecology, César Capinha, Ana Ceia-Hasse, Andrew M. Kramer, Christiaan Meijer

Integrative Biology Faculty and Staff Publications

Temporal data is ubiquitous in ecology and ecologists often face the challenge of accurately differentiating these data into predefined classes, such as biological entities or ecological states. The usual approach consists of transforming the time series into user-defined features and then using these features as predictors in conventional statistical or machine learning models. Here we suggest the use of deep learning models as an alternative to this approach. Recent deep learning techniques can perform the classification directly from the time series, eliminating subjective and resource-consuming data transformation steps, and potentially improving classification results. We describe some of the deep learning …

Long-Term Changes In Juvenile Green Turtle Abundance And Foraging Ecology In The Indian River Lagoon, Florida, Christopher Long

Electronic Theses and Dissertations, 2020-2023

Marine turtles are distributed in temperate, sub-tropical and tropical waters and beaches worldwide, often in areas heavily impacted by humans. Although there are many threats to marine turtle populations, the growing threats of nutrient pollution and harmful algal blooms are relatively understudied despite their widespread impacts on coastal marine ecosystems that marine turtles depend on. By studying juvenile green turtles (Chelonia mydas) in the Indian River Lagoon, Florida, where nutrient pollution and HABs are a widespread and longstanding issue, I aimed to conduct a case study of how these threats may affect this federally Threatened species. In Chapter 2, I …

An Evaluation Of Antiparasitics Used To Treat The Sea Louse, Caligus Rogercresseyi, In Chilean Salmonid Farming, Dayla Woller

Scripps Senior Theses

Sea lice pose a global threat to open-net salmonid farming. As Chile is the second largest Atlantic salmon producer, they have a vested interest in mitigating infestations and increasing production. However, this will not be possible without 1) increased research on the underlying mechanisms behind sea lice infestations that will inform 2) governmental regulation of salmon farming. This increase in research and governmental regulation will only be possible through transparency on pesticide practices and sea louse monitoring data from companies operating in Chile. Taking these steps would bring Chile closer to being the largest producer of Atlantic salmon and an …

Lepidopteran Granivory Reduces Seed Counts In A Rare Species Of Riparian Scour Prairies, Cheyenne Moore, Angela J. Mcdonnell, Scott Schuette, Christopher T. Martine

Faculty Journal Articles

In Pennsylvania Baptisia australis var. australis is found along only four waterways: the Allegheny River, Youghiogheny River, Clarion River, and Red Bank Creek. Because of its limited distribution and small number of extant populations, the species is considered state-threatened in Pennsylvania. In addition, the riparian prairie habitat that Pennsylvania Baptisia australis var. australis is restricted to is also in decline and considered vulnerable. Because of these conservation concerns, insights into the natural history of the taxon in the state is valuable and will inform conservation efforts. Field surveys and fruit collections along the Allegheny River and herbarium collections were used …

A Decision Tool To Identify Population Management Strategies For Common Ravens And Other Avian Predators, Andrea F. Currylow, Brenda J. Hanley, Kerry L. Holcomb, Timothy Shields, Stephen Boland, William I. Boarman, Mercy Vaughn

Human–Wildlife Interactions

Some avian species have developed the capacity to leverage resource subsidies associated with human manipulated landscapes to increase population densities in habitats with naturally low carrying capacities. Elevated corvid densities and new territory establishment have led to an unsustainable increase in depredation pressure on sympatric native wildlife prey populations as well as in crop damage. Yet, subsidized predator removal programs aimed at reducing densities are likely most effective longer-term when conducted in tandem with subsidy control, habitat management, and robust assessment monitoring programs. We developed decision support software that leverages stage structured Lefkovitch population matrices to compare and identify treatment …

Leveraging Transcriptomic Approaches To Identify Differences In Genetic Programming Driving Two Distinct Wound Healing Mechanisms, Regeneration And Fibrosis, In Acomys And Mus, Shishir K. Biswas

Theses and Dissertations--Biology

Why can some animals and others cannot? This fundamental question has fueled scientists studying regeneration for hundreds of years since early observations in crayfish, salamanders and many other organisms. While most contemporary work in regeneration is done in a handful of species including salamanders, zebrafish and flatforms, these organisms lack a closely-related, non-regenerating sister species from which unique genetic differences can be identified. Additionally, while much has been learned from these organisms, they do not share fundamental biological traits with mammals (endothermy, metabolism and immune system) which limits the ability to translate this research for clinical medicine. To this end, …

Investigation Into The Genetic Provenance Of Three Rare Plants With East-West Disjunction Patterns In Pennsylvania., Scott Schuette, Christopher T. Martine

Other Faculty Research and Publications

Rare plant conservation relies on an understanding of the natural history, biology and ecology, and real and potential threats to their populations to inform state regulations that serve to protect the species from extirpation. This work often involves extensive field surveys over several years to determine population sizes and whether those populations are seeing reductions in number of individuals necessary to maintain the genetic diversity within and between those populations. Species and populations with high genetic diversity are better equipped to withstand sudden changes to their habitats that derive from land use changes and changing climate. There are a variety …

The Mechanisms And Consequences Of Shrub Encroachment On The Virginia Barrier Islands, Lauren K. Wood

Theses and Dissertations

Shrub encroachment is a global phenomenon driven by direct and indirect anthropogenic influence which alters plant communities and ecosystem function. Many studies have investigated drivers and consequences of woody plant establishment, but mesic landscapes are underrepresented in the literature. My objective was to assess the mechanisms of Morella cerifera encroachment into coastal mesic grassland, the potential for self-reinforcement, and consequences on community composition, nutrients, and landscape productivity. I studied temperature and water microclimate modification by Morella cerifera presence and removal to understand ecosystem engineering and community composition changes. Additionally, I examined the influence of shrubs on surrounding grassland species traits …

Composition And Homology In The Taxonomic Classification Of Escherichia Coli, Tanya Irani

Theses and Dissertations (Comprehensive)

As new techniques have been introduced, specifically the possibility of complete genome sequencing, better methods of defining bacterial species have also been proposed. One of the most recently proposed methods, using bioinformatic techniques, is to calculate the average nucleotide identity (ANI) between the homologous genome segments of different isolates. Another method for species discrimination that has been tested successfully is the similarity of DNA compositional signatures. However, in a recent update, DNA signatures split the available Escherichia coli complete genomes into three groups. To check if this result was consistent with such genomes belonging to different species, we tested methods …

Nitrogen Fixation: A Poorly Understood Process Along The Freshwater-Marine Continuum, Amy M. Marcarelli, Robinson W. Fulweiler, J. Thad Scott

Michigan Tech Publications, Part 1

N2 fixation is a major component of the global N cycle and has been extensively studied in open-ocean and terrestrial ecosystems. Yet rates and ecological dynamics remain virtually unknown for the inland and coastal aquatic ecosystems (lakes, wetlands, rivers, streams, and estuaries) that connect terrestrial and marine biomes. This is due to the diversity of these habitats as well as the traditional paradigm that N2 fixation rates were low to nonexistent, and therefore not important, in these ecosystems. We identify three major research themes to advance understanding of aquatic N2 fixation: (1) the biological diversity of diazotrophs and variability of …

A Bioinformatic Analysis Of The Biosynthesis Of Carotenoids In The Copepod Tigriopus Californicus, Anchalya Balasubramaniam

Theses and Dissertations (Comprehensive)

Abstract

Biological pigments, also called biochromes, are coloured compounds which are displayed by a variety of life forms, including animals, due to selective colour absorption. The combination of light absorption and reflection enables each pigment to portray a distinct colour which results in the broad spectrum of colours we observe in our surroundings. Carotenoids are a large group of yellow, orange, and red biological pigments found in living organisms. Our current biomolecular knowledge of carotenoids is heavily derived from studying the pathway in photosynthetic prokaryotes, bacteria, fungi, and plants. Carotenoid pigments are exceptionally multifunctional as they act as photo-protectors against …

Investigating The Antitumor Effects Of A Dsrna-Nanoparticle Complex In An In Vitro Ovarian Cancer Model, Aaron Lewis

Theses and Dissertations (Comprehensive)

An estimated 1 in 70 women will be diagnosed with ovarian cancer in their lifetime. Despite advanced detection and treatment methods, it remains a silent killer with an expected survival rate of 50%. A developing method in cancer treatment is the use of compounds that stimulate the immune system to aid in the body's fight against the disease. This project focused on the use of the potent immune stimulant double-stranded RNA (dsRNA), commercially available as polyinosinic:polycytidylic acid, poly(I:C), to induce cytotoxicity in two ovarian cancer cell lines; SKOV-3 and OVCAR-3. Some challenges exist with the delivery of dsRNA due to …

Dsrna Conjugated Phytoglycogen Nanoparticles As An Innate Immune Stimulant To Prevent Viral Infections In Rainbow Trout And Atlantic Salmon, Kayla Samms

Theses and Dissertations (Comprehensive)

Salmonids are one of the world’s most farmed fish species. These aquatic vertebrate species rely heavily on their innate immune responses to defend themselves against invading pathogens. Although commercial vaccines are available against some viral and bacterial pathogens affecting salmonids, their protective efficacy is variable. Using a prophylactic inducer of local and systemic innate immune responses could have significant implications in salmonid aquaculture. A potent inducer of the innate immune response in fish is double-stranded RNA (dsRNA), a molecule that all viruses make during their replicative cycle. Polyinosinic:polycytidylic acid (polyI:C) is a synthetic dsRNA commonly used to induce type I …

Wildlife Forage Recovery Following Boreal Wildfire, Alexis Jorgensen, Jennifer Baltzer

Theses and Dissertations (Comprehensive)

Climate change is altering the boreal wildfire regime through increases in the extent and severity of burning and reductions in fire return intervals. These changes can alter the regeneration trajectory of canopy species and ground vegetation, with implications for wildlife habitat. There is some uncertainty about the timelines of when different animal species will use burned areas as their preferred forage taxa recover following fire, and how such recovery is mediated by environmental factors. Here, we aim to address these knowledge gaps through the following questions: 1) What are the main forage types consumed by boreal wildlife and how much …

How Does Direct Volunteering Affect The Onset Of Alzheimer’S Dementia In Elderly Patients With Preexisting Cardiac Comorbidities?, Sneha R. Gade

Undergraduate Research Posters

Alzheimer's Dementia is a disease that affects memory, thinking, and behavior. The Alzheimer's Association reported Alzheimer's as the sixth-leading cause of death in the United States, and a cure does not exist. Donepezil, a cholinesterase inhibitor drug, is frequently prescribed to treat Alzheimer's disease. Still, patients must continuously take the medication for years to receive any measurable improvement in quality of life after developing Alzheimer's Dementia. Further, research shows prolonged use of Donepezil can lead to other health problems, such as QTc prolongation, for cardiac disease patients. Therefore, the effectiveness of non-drug interventions, such as volunteering to prevent the disease's …

Temporal Thermal Heterogeneity As A Potential Evolutionary Driver Of Thermoregulation In The Generalist Grasshopper Melanoplus Differentialis From Five Sites In The Midwestern And Southern United States, Devin Preston

LSU New Orleans Theses and Dissertations

Environmental drivers of the evolution of thermoregulation, especially thermal fluctuation, have been incompletely investigated. I used a combination of field and laboratory studies to determine if thermal heterogeneity selected for greater thermal tolerance or ability to thermoregulate in a generalist grasshopper by testing upper and lower thermal limits, behavioral thermoregulation, and acclimation. I measured these from individuals originating from sites that had varying degrees of thermal fluctuation over time. I found partial support for thermal heterogeneity as an evolutionary driver of thermal tolerance, behavioral thermoregulation, and acclimation. However, whether or not differences in these metrics conformed to my hypothesis varied …