26800 research outputs found
Sort by
GROWTH, SURVIVAL, AND HERBIVORE BROWSING PREFERENCES OF SEEDLINGS PLANTED IN BLACK ASH STANDS OF NORTHERN MICHIGAN AND WISCONSIN
No full textTo mitigate the cascading ecosystem impacts of widespread black ash (Fraxinus nigra) mortality due to the invasive emerald ash borer (Agrilus planipennis), forested wetlands can be underplanted with seedlings of alternative species that later repopulate the vacated canopy. This study examined the survival and growth, factors influencing performance, and herbivore browsing preferences of seven non-ash species planted in northern Michigan and Wisconsin. The best performing species were river birch (Betula nigra), silver maple (Acer saccharinum), and swamp white oak (Quercus bicolor). The worst performing species was black spruce (Picea mariana). While maples and oaks were the most-consumed species by herbivores, browsing did not significantly affect overall performance of these species. Seedlings planted higher above the water table exhibited greatly increased chances of survival. Overall, this study identified several species that can survive and thrive under wet forest conditions, providing foresters greater opportunities for forest restoration in ash-dominated stands
Novel Statistical Methods for Multiple Phenotype and Gene Based Association Tests
No full textIn chapter one, we proposed a novel multiple phenotype association test methods to test the association between multiple phenotypes and a SNP. Genome-wide association studies (GWAS) have identified many strongly associated genetic variants with phenotypes and have greatly enhanced our understanding of the genetic architecture of complex phenotypes and diseases. In this study, we proposed a new method to construct a Phenotype-Phenotype Network using the sparse Gaussian Graphical Model (sGGM) based on GWAS summary statistics. This approach will isolate the direct relationship between phenotypes, making it easier to identify clusters of phenotypes conditional on other phenotypes that reflect the more meaningful biological or functional connections. Then we applied the community detection method to partition phenotypes into disjoint modules based on the partial correlation matrix of phenotypes and applied multiple phenotype test methods on each of the clusters and combined p-values to get the final P-value. The comprehensive simulation study showed that our method can control the type I error rate effectively and has the highest power compared to the other methods we compared. Application of this method to the GWAS summary statistics of 92 phenotypes from UK-Biobank data has identified higher number of significant SNPs than other methods. Downstream analysis of the significant SNPs shows the functional and biological importance.
The complex traits are often the result of the combined effect of multiple genetic variants, each with a small individual effect and these small effects may be missed due to stringent p-value thresholds in GWAS studies. Gene-based association tests combine the effects of multiple variants within a gene and provide more interpretable associations and enable researchers to detect associations that might be missed in single-SNP analyses. Gene-based tests increase statistical power by considering the cumulative effect of several variants in a gene, reducing the multiple testing burden and prioritizing variants based on their likelihood of contributing to the phenotype. The power of the gene-based test depended on the genetic architecture of the trait and the weight of SNPs in a gene. The genetic architecture of the complex trait is not known in advance and the power of the gene-based association tests decreases if the weight is mis-specified. The equal weighting assumption implies that all SNPs within the gene have an equal effect on the trait. On the other hand, minor allele frequency (MAF)-based weights, such as inverse or beta distribution of MAF, upweight rare variants under the assumption that they are more likely to be deleterious. These weighting schemes also ignore the underlying biological function of SNPs, which, if integrated, can enhance the power of gene-based tests by prioritizing functionally relevant SNPs. In this study, we propose a novel gene-based test approach (H2-Gene) that weights SNPs according to their heritability estimated using multiple functional annotations. The comprehensive simulation studies showed that this approach can control the false discovery rate very well and yield the highest power compared to MAF weight and equal weight. When applied to GWAS summary statistics for three different traits: Schizophrenia, Bipolar Disorder, and Attention Deficit Hyperactivity Disorder, H2-Gene identified a greater number of significant genes relative to equal and MAF-based weighting schemes. The downstream analysis of the significant genes showed biological and functional importance of the genes
IDENTIFYING PATTERNS IN AVIAN PLASMA METABOLITES FOR USE IN COMPARING BETWEEN-SITE DIFFERENCES IN HABITAT QUALITY
No full textHabitat loss and degradation are especially important considerations for migratory songbirds, since they depend on a variety of distinct habitats within different life phases (such as breeding, wintering, and migration) that are widely dispersed across a spatial scale. Additionally, abundance of migratory songbirds has been in severe decline, making habitat maintenance and recovery of particular interest as a research avenue. Identifying high quality habitats across a bird’s annual range is therefore of increasing importance and represents fundamental information for guiding management decision making. Free-floating metabolites found in blood are frequently used to determine short-term mass changes in birds and have been utilized to intrinsically reveal variation in habitat quality across dissimilar habitats. We collected triglyceride and β-hydroxybutyrate concentrations from songbird blood samples across seven different sites located in Michigan’s Keweenaw Peninsula across the breeding season and fall migration to better understand variation in metabolite values across life phases. Consistent with other similar studies, TRIG and BUTY values were negatively correlated, with a stronger relationship observed during fall migration. Sex-based differences in TRIG was apparent during the breeding season, with females having higher concentrations than males (p \u3c 0.001), which is consistent with lipid allocation for egg production. Mixed linear modeling indicated that the highest variation in metabolite values was due to site, net, species, and individual differences, rather than direct relationships between TRIG and BUTY. Interpretation of TRIG and BUTY combinations during migration revealed distinct metabolic states, including energy gain, deficit, and equilibrium. Notably, high TRIG in individuals with no visible fat reserves likely reflected dietary refueling rather than fat mobilization, underscoring the importance of interpreting metabolic values in context as energetic indicators. Our results highlight the shifts in metabolite usage across life stages and suggest that variation between groups across space and time should be accounted for in future research
MICROBIAL HYDROCARBON DEGRADATION AND CARBON CYCLING IN MARINE AND FRESHWATER SEDIMENTS: A COMPARATIVE METAGENOMIC ANALYSIS
No full textMicrobes are essential in keeping our environment balanced, especially by fixing carbon dioxide and breaking down hydrocarbons in aquatic sediments. This thesis explores how microbial communities perform these tasks in different sediment environments, including marine areas (Caspian Sea and Mediterranean Sea) and freshwater ecosystems (Great Lakes). The first part of this thesis reviews current knowledge about how microbes degrade hydrocarbons. It highlights the importance of different pathways microbes use depending on oxygen availability. The second part provides a detailed study of the microbial communities found in marine sediments from the Caspian Sea and Mediterranean Sea, showing that microbes in these deeper, oxygen-limited sediments mainly use anaerobic (oxygen-free) pathways for degrading hydrocarbons.
In the third part, another review focuses on hydrocarbon degradation mechanisms in freshwater (Great Lakes). The fourth part investigates microbial carbon fixation, the process microbes use to convert carbon dioxide into organic matter in sediments from marine and freshwater environments. The results reveal that marine microbes, especially in deeper sediments, predominantly rely on anaerobic carbon fixation pathways like the Wood–Ljungdahl pathway and reverse TCA cycle. In contrast, freshwater sediments from the Great Lakes show greater diversity, with microbes using aerobic (oxygen-requiring) pathways such as glycine and hydroxypropionate cycles due to higher oxygen availability. An important discovery of this thesis is the metabolic flexibility of certain microbes, known as facultative autotrophy. Based on environmental conditions, these microbes can switch between using inorganic carbon (carbon dioxide) and organic carbon (such as hydrocarbons). This ability helps them survive and adapt to changing sediment environments. Additionally, this research found that carbon fixation pathways are present in various microbes, including many previously not recognized for this ability, emphasizing their importance across microbial life.
Overall, this work significantly improves our understanding of how sediment microbes adapt their metabolic strategies according to environmental conditions like oxygen levels and sediment depth. These insights help us predict how microbial communities respond to environmental changes, making this knowledge valuable for future environmental management, carbon sequestration strategies, and pollution cleanup efforts
Tailored control of evaporation flux for uniform coffee-ring patterns in multiple nanofluid droplets
No full textThe present study proposes a straightforward approach to achieving consistent particle deposition patterns in multiple nanofluid droplet evaporation systems by strategically placing de-ionized (DI) water droplets without nanoparticles at both ends of the nanofluid droplet array. A proof-of-concept investigation was conducted through a combination of experimental and numerical analyses for a five-droplet evaporation system, consisting of three nanofluid droplets and two DI water droplets, to control vapor shielding effects. This study also examined contact-line dynamics, vapor concentration distribution, and evaporation flux during the evaporation process. The findings revealed inconsistent coffee-ring patterns formed by the three nanofluid droplets, without the addition of DI water droplets. In contrast, when DI water droplets were placed at both ends of the array, we were able to fabricate nearly consistent coffee-ring patterns. Specifically, the consistency of the coffee-ring patterns improved with an increase in DI water droplet volume, achieved by tailoring the evaporation flux near the droplet edges. Increased DI water volume also extended the total evaporation time due to enhanced vapor accumulation. Moreover, DI water droplets exhibited pinning-depinning behavior during evaporation, whereas nanofluid droplets remained pinned for longer durations as particle aggregation delayed the depinning process
Investigating the Inequality of Phase Change Coefficients Using ISS Experimental Data
No full textKinetic theory is a popular approach to model liquid-vapor phase change but accurate determination of evaporation and condensation coefficients remains a challenge. Reported values of coefficients vary by several orders of magnitude. For simplicity and convenience evaporation and condensation coefficients are assumed to be equal though there is little physical evidence to support this. This study presents a novel methodology to test this assumption using data from Constrained Vapor Bubble (CVB) experiments conducted on the International Space Station (ISS). The experiments consist of a quartz cuvette that is partially filled with n-pentane; heated and cooled at opposite ends to induce simultaneous evaporation and condensation around a central bubble. Data obtained from the NASA Physical Sciences Informatics (PSI) database enabled a three-dimensional reconstruction of the liquid–vapor interface. The net mass flux over the vapor bubble surface is zero at steady operation, providing a closure relationship for simultaneous and independent calculation of both evaporation and condensation coefficients. The resulting coefficient values are within 1% of each other but are not equal. The two coefficients are also within 2% of those predicted using transition state theory. When the evaporation and condensation coefficients are forced to be equal, the deviation from transition state theory is approximately 60%. This deviation monotonically increases with increasing rates of evaporation/condensation due to a systemic under-prediction of the bubble surface area. The agreement between derived coefficients and those predicted by transition state theory is maintained when the bubble surface area is corrected to account for Marangoni-induced interfacial instabilities
Machine Learning-based Prediction of Temporal Velocity-Informatics (TVI) Variables for Accelerated Characterization of Intracranial Aneurysms’ Rupture Status
No full textAbstract: Temporal velocity-informatics (TVI) is a novel technique utilizing spatial analysis of time-resolved 3D velocity fields to quantify flow disturbance in vascular aneurysms. Although it can improve the characterization of intracranial aneurysms’ (IA) rupture status, calculation of time-resolved 3D velocity fields using computational fluid dynamics (CFD) simulations limits its clinical translation. This study aims to test the feasibility of using IA’s geometrical information in conjunction with machine learning (ML)-based regression methods to predict TVI parameters. The effectiveness of these ML-predicted TVI parameters in predicting IA’s rupture status was evaluated using one hundred twelve IAs with known rupture status. We found that predicting the IA’s rupture status using predicted TVI can achieve an AUC of 0.88, and a total accuracy of 81.6%. Also, We found that the consistency between ML-predicted TVI variables and estimated TVI metrics calculated from CFD-simulated velocity data was higher than our ability to predict wall shear stress-based metrics. Graphic abstract: (Figure presented.
Multitask Learning-Based Approaches for Protein Function Prediction
No full textAdvancements in sequencing technologies have resulted in a massive growth in the number of sequences available. Only a small fraction of the proteins in UniProtKB have been functionally annotated. Understanding the roles and studying the mechanisms of newly discovered proteins is one of the most important biological problems in the post-genomic era. To address the sequence-function gap many computational methods have been developed. This chapter reviews Multitask Learning (MTL)-based approaches for protein function prediction, highlighting its potential to enhance both predictive accuracy and computational efficiency in bioinformatics. MTL utilizes shared representations to leverage common information across related tasks, improving predictive performance. Key findings reveal that MTL improves predictive performance by integrating shared features across related tasks
Near Memory Accelerators for Approximate Nearest Neighbor Search
No full textApproximate Nearest Neighbor (ANN) search has become a critical component in modern retrieval systems, supporting applications such as recommendation engines, semantic search, and large language models. Despite the effectiveness of embedding-based methods, performing ANN search over billion-scale, high-dimensional datasets remains computationally intensive. To address these challenges, this dissertation explores three system-level accelerator designs that improve the performance, scalability, and efficiency of ANN search. Our first work focuses on graph-based ANN search and addresses key bottlenecks in memory access and distance computation. We propose a near-memory accelerator that performs distance calculations within DRAM and transfers only compact results to the compute engine. To further enhance throughput, we propose parallel vertex expansion with bounded staleness and a prefetching strategy that improves memory access efficiency. This design achieves a 93.9% improvement in throughput, and further optimizations yield a 2.22× speedup over a state-of-the-art graph-based accelerator. Our second work targets Inverted File Product Quantization (IVFPQ)-based ANN search. Existing PQ accelerators suffer from excessive data movement and inefficient memory usage. We design a near-memory accelerator with in-DRAM distance computation, along with memory-aware cluster placement to balance workload. We also propose distance filters to eliminate non-contributing values and an asymmetric quantization method to reduce LUT storage without recall degradation. This design delivers an 11.3× throughput improvement and a 91.1% reduction in memory traffic compared to the ANNA accelerator. Our third work addresses scalability limitations by leveraging Compute Express Link (CXL) to access disaggregated memory. We propose CXL-ANNX, a distributed near-memory accelerator that executes sub-queries across multiple remote DRAM modules connected via CXL. To mitigate CXL’s latency and bandwidth constraints, we develop several system-level techniques: Termination Execution of Unnecessary Sub-queries, Speculative Search, and Memory-Aware Cluster Placement. Additionally, we integrate a lightweight learning-based Early Exit mechanism that reduces remote memory accesses by dynamically predicting termination points. Across four real-world datasets, CXL-ANNX with ALL-OPT achieves 13.1× throughput improvement over DiskANN on billion-scale datasets. Together, these contributions demonstrate a scalable and efficient architecture stack for high-performance ANN search, spanning graph- and IVFPQ-based methods, near-memory computing, and disaggregated memory systems
