Browsing by Subject "principal component analysis"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Combining Multivariate Statistical Methods and Spatial Analysis to Characterize Water Quality Conditions in the White River Basin, Indiana, U.S.A.
    (2011-02-25) Gamble, Andrew Stephan; Babbar-Sebens, Meghna; Tedesco, Lenore P.; Peng, Hanxiang
    This research performs a comparative study of techniques for combining spatial data and multivariate statistical methods for characterizing water quality conditions in a river basin. The study has been performed on the White River basin in central Indiana, and uses sixteen physical and chemical water quality parameters collected from 44 different monitoring sites, along with various spatial data related to land use – land cover, soil characteristics, terrain characteristics, eco-regions, etc. Various parameters related to the spatial data were analyzed using ArcHydro tools and were included in the multivariate analysis methods for the purpose of creating classification equations that relate spatial and spatio-temporal attributes of the watershed to water quality data at monitoring stations. The study compares the use of various statistical estimates (mean, geometric mean, trimmed mean, and median) of monitored water quality variables to represent annual and seasonal water quality conditions. The relationship between these estimates and the spatial data is then modeled via linear and non-linear multivariate methods. The linear statistical multivariate method uses a combination of principal component analysis, cluster analysis, and discriminant analysis, whereas the non-linear multivariate method uses a combination of Kohonen Self-Organizing Maps, Cluster Analysis, and Support Vector Machines. The final models were tested with recent and independent data collected from stations in the Eagle Creek watershed, within the White River basin. In 6 out of 20 models the Support Vector Machine more accurately classified the Eagle Creek stations, and in 2 out of 20 models the Linear Discriminant Analysis model achieved better results. Neither the linear or non-linear models had an apparent advantage for the remaining 12 models. This research provides an insight into the variability and uncertainty in the interpretation of the various statistical estimates and statistical models, when water quality monitoring data is combined with spatial data for characterizing general spatial and spatio-temporal trends.
  • Thumbnail Image
    Item
    Enhancement of osteoblastogenesis and suppression of osteoclastogenesis by inhibition of de-phosphorylation of eukaryotic translation initiation factor 2 alpha
    (Smart Science and Technology, LLC, 2015) Hamamura, Kazunori; Chen, Andy; Yokota, Hiroki; Department of Anatomy and Cell Biology, IU School of Medicine
    The phosphorylation of eukaryotic translation initiation factor 2 alpha (eIF2α) is activated in response to various stresses such as viral infection, nutrient deprivation, and stress to the endoplasmic reticulum. Severe stress to the endoplasmic reticulum, for instance, induces an apoptotic pathway, while mild stress, on the contrary, leads to a pro-survival pathway. Little has been known about the elaborate role of eIF2α phosphorylation in the development of bone-forming osteoblasts and bone-resorbing osteoclasts. Using salubrinal and guanabenz as inhibitors of the de-phosphorylation of eIF2α, we have recently reported that the phosphorylation of eIF2α significantly alters fates of both osteoblasts and osteoclasts. Based on our recent findings, we review in this research highlight the potential mechanisms of the enhancement of osteoblastogenesis and the suppression of osteoclastogenesis through the elevated level of phosphorylated eIF2α.
  • Thumbnail Image
    Item
    Improved Analysis and Visualization of Community Indicators and Indices
    (Office of the Vice Chancellor for Research, 2013-04-05) Farah, Christopher; Kandris, Sharon; Frederickson, Karen Comer
    The analysis and interpretation of community indicators has been widely conducted to understand community trends, and subsequently, to support program planning, public policy initiatives, and target geographic regions for research. Given the importance of the outcomes, selecting good indicators is key, and usually a balance of stakeholder input and analytical evaluation. The most common analysis method used to evaluate a set of indicators is principal component analysis (PCA), a linear multivariate analysis method. However, the assumptions of PCA may be too restrictive and consequently, the analysis may fail to provide a sound evaluation of the set of indicators. In response to this shortcoming, we paired PCA with an unsupervised, non-linear multivariate method, known as self-organizing maps (SOMs), to analyze a set of indicators focused on population trends in education, income, employment, among others, at both the county level and the census tract level. The joint results were used to: exclude or include indicators from the indicator set, determine the latent primary dimensions of the dataset, identify peer counties and census tracts (relative to Indiana counties / census tracts), identify associations among different indicators at different geographic scales, identify temporal changes in the value of indicators, and develop one or more indices to describe socio-economic conditions of communities. Outcomes are presented geographically, topologically, tabularly, and graphically, offering different mechanisms of understanding and interpreting the analysis results. A goal of this project is to provide a web-based interface for researchers and community stakeholders to identify and evaluate candidate sets of community indicators, potentially accelerating sound public policy decisions and public health research.
  • Thumbnail Image
    Item
    Predicting locations for urban tree planting
    (2014) King, Steven M.; Johnson, Daniel P. (Daniel Patrick), 1971-; Bein, Frederick L. (Frederick Louis), 1943-; Lulla, Vijay O.
    The purpose of this study was to locate the most suitable blocks to plant trees within Indianapolis, Indiana’s Near Eastside Community (NESCO). LiDAR data were utilized, with 1.0 meter average post spacing, captured by the Indiana Statewide Imagery and LiDAR Program from March 13, 2011 to April 30, 2012, to conduct a covertype classification and identify blocks that have low canopies, high impervious surfaces and high surface temperatures. Tree plantings in these blocks can help mitigate the effects of the urban heat island effect. Using 2010 U.S. Census demographic data and the principal component analysis, block groups with high social vulnerability were determined, and tree plantings in these locations could help reduce mortality from extreme heat events. This study also determined high and low priority plantable space in order to emphasize plantable spaces with the potential to shade buildings; this can reduce cooling costs and the urban heat island, and it can maximize the potential of each planted tree.
  • Thumbnail Image
    Item
    Towards Subject and Diagnostic Identifiability in the Alzheimer’s Disease Spectrum Based on Functional Connectomes
    (Springer, 2018-01) Svaldi, Diana O.; Goñi, Joaquín; Sanjay, Apoorva Bharthur; Amico, Enrico; Risacher, Shannon L.; West, John D.; Dzemidzic, Mario; Saykin, Andrew; Apostolova, Liana; Neurology, School of Medicine
    Alzheimer’s disease (AD) is the only major cause of mortality in the world without an effective disease modifying treatment. Evidence supporting the so called “disconnection hypothesis” suggests that functional connectivity biomarkers may have clinical potential for early detection of AD. However, known issues with low test-retest reliability and signal to noise in functional connectivity may prevent accuracy and subsequent predictive capacity. We validate the utility of a novel principal component based diagnostic identifiability framework to increase separation in functional connectivity across the Alzheimer’s spectrum by identifying and reconstructing FC using only AD sensitive components or connectivity modes. We show that this framework (1) increases test-retest correspondence and (2) allows for better separation, in functional connectivity, of diagnostic groups both at the whole brain and individual resting state network level. Finally, we evaluate a posteriori the association between connectivity mode weights with longitudinal neurocognitive outcomes.