Analysis Of Multiple Data Types Research Articles

Interrelated impacts of climate and land-use change on a widespread waterbird.

Together climate and land-use change play a crucial role in determining species distribution and abundance, but measuring the simultaneous impacts of these processes on current and future population trajectories is challenging due to time lags, interactive effects and data limitations. Most approaches that relate multiple global change drivers to population changes have been based on occurrence or count data alone. We leveraged three long-term (1995-2019) datasets to develop a coupled integrated population model-Bayesian population viability analysis (IPM-BPVA) to project future survival and reproductive success for common loons Gavia immer in northern Wisconsin, USA, by explicitly linking vital rates to changes in climate and land use. The winter North Atlantic Oscillation (NAO), a broad-scale climate index, immediately preceding the breeding season and annual changes in developed land cover within breeding areas both had strongly negative influences on adult survival. Local summer rainfall was negatively related to fecundity, though this relationship was mediated by a lagged interaction with the winter NAO, suggesting a compensatory population-level response to climate variability. We compared population viability under 12 future scenarios of annual land-use change, precipitation and NAO conditions. Under all scenarios, the loon population was expected to decline, yet the steepest declines were projected under positive NAO trends, as anticipated with ongoing climate change. Thus, loons breeding in the northern United States are likely to remain affected by climatic processes occurring thousands of miles away in the North Atlantic during the non-breeding period of the annual cycle. Our results reveal that climate and land-use changes are differentially contributing to loon population declines along the southern edge of their breeding range and will continue to do so despite natural compensatory responses. We also demonstrate that concurrent analysis of multiple data types facilitates deeper understanding of the ecological implications of anthropogenic-induced change occurring at multiple spatial scales. Our modelling approach can be used to project demographic responses of populations to varying environmental conditions while accounting for multiple sources of uncertainty, an increasingly pressing need in the face of unprecedented global change.

Coupled co-clustering-based unsupervised transfer learning for the integrative analysis of single-cell genomic data.

Unsupervised methods, such as clustering methods, are essential to the analysis of single-cell genomic data. The most current clustering methods are designed for one data type only, such as single-cell RNA sequencing (scRNA-seq), single-cell ATAC sequencing (scATAC-seq) or sc-methylation data alone, and a few are developed for the integrative analysis of multiple data types. The integrative analysis of multimodal single-cell genomic data sets leverages the power in multiple data sets and can deepen the biological insight. In this paper, we propose a coupled co-clustering-based unsupervised transfer learning algorithm (coupleCoC) for the integrative analysis of multimodal single-cell data. Our proposed coupleCoC builds upon the information theoretic co-clustering framework. In co-clustering, both the cells and the genomic features are simultaneously clustered. Clustering similar genomic features reduces the noise in single-cell data and facilitates transfer of knowledge across single-cell datasets. We applied coupleCoC for the integrative analysis of scATAC-seq and scRNA-seq data, sc-methylation and scRNA-seq data and scRNA-seq data from mouse and human. We demonstrate that coupleCoC improves the overall clustering performance and matches the cell subpopulations across multimodal single-cell genomic datasets. Our method coupleCoC is also computationally efficient and can scale up to large datasets. Availability: The software and datasets are available at https://github.com/cuhklinlab/coupleCoC.

Interrelated ecological impacts of climate change on an apex predator.

Climate change has broad ecological implications for species that rely on sensitive habitats. For some top predators, loss of habitat is expected to lead to cascading behavioral, nutritional, and reproductive changes that ultimately accelerate population declines. In the case of the polar bear (Ursus maritimus), declining Arctic sea ice reduces access to prey and lengthens seasonal fasting periods. We used a novel combination of physical capture, biopsy darting, and visual aerial observation data to project reproductive performance for polar bears by linking sea ice loss to changes in habitat use, body condition (i.e., fatness), and cub production. Satellite telemetry data from 43 (1991–1997) and 38 (2009–2015) adult female polar bears in the Baffin Bay subpopulation showed that bears now spend an additional 30 d on land (90 d in total) in the 2000s compared to the 1990s, a change closely correlated with changes in spring sea ice breakup and fall sea ice formation. Body condition declined for all sex, age, and reproductive classes and was positively correlated with sea ice availability in the current and previous year. Furthermore, cub litter size was positively correlated with maternal condition and spring breakup date (i.e., later breakup leading to larger litters), and negatively correlated with the duration of the ice‐free period (i.e., longer ice‐free periods leading to smaller litters). Based on these relationships, we projected reproductive performance three polar bear generations into the future (approximately 35 yr). Results indicate that two‐cub litters, previously the norm, could largely disappear from Baffin Bay as sea ice loss continues. Our findings demonstrate how concurrent analysis of multiple data types collected over long periods from polar bears can provide a mechanistic understanding of the ecological implications of climate change. This information is needed for long‐term conservation planning, which includes quantitative harvest risk assessments that incorporate estimated or assumed trends in future environmental carrying capacity.

Physically Consistent Modeling of Dike‐Induced Deformation and Seismicity: Application to the 2014 Bárðarbunga Dike, Iceland

AbstractDike intrusions are often associated with surface deformation and propagating swarms of earthquakes. These are understood to be manifestations of the same underlying physical process, although rarely modeled as such. We construct a physics‐based model of the 2014 Bárðarbunga dike, by far the best observed large dike ( 0.5 km ) to date. We constrain the background stress state by the total dike deformation, the time‐dependent dike pressure from continuous GPS and the extent of the seismic swarm, and the spatial dependence of frictional properties via the space‐time evolution of seismicity. We find that the geodetic and earthquake data can be reconciled with a self‐consistent set of parameters. The complex spatial and temporal evolution of the Bárðarbunga seismicity can be explained by dike‐induced elastic stress changes on preexisting faults, constrained by observed focal mechanisms. In particular, the model captures the segmentation of seismicity, where only the newest dike segment is seismically active. Our results indicate that many features of the seismicity result from the interplay between time‐dependent magma pressure within the dike and stress memory effects. The spatial variability in seismicity requires heterogeneity in frictional properties and/or local initial stresses. Modeling suggests that the dike pressure drops during rapid advances and increases during pauses, which primarily causes the segmentation of the seismicity. Joint analysis of multiple data types could potentially lead to improved, physics‐based eruption forecasts.

Non-parametric combination analysis of multiple data types enables detection of novel regulatory mechanisms in T cells of multiple sclerosis patients

Multiple Sclerosis (MS) is an autoimmune disease of the central nervous system with prominent neurodegenerative components. The triggering and progression of MS is associated with transcriptional and epigenetic alterations in several tissues, including peripheral blood. The combined influence of transcriptional and epigenetic changes associated with MS has not been assessed in the same individuals. Here we generated paired transcriptomic (RNA-seq) and DNA methylation (Illumina 450 K array) profiles of CD4+ and CD8+ T cells (CD4, CD8), using clinically accessible blood from healthy donors and MS patients in the initial relapsing-remitting and subsequent secondary-progressive stage. By integrating the output of a differential expression test with a permutation-based non-parametric combination methodology, we identified 149 differentially expressed (DE) genes in both CD4 and CD8 cells collected from MS patients. Moreover, by leveraging the methylation-dependent regulation of gene expression, we identified the gene SH3YL1, which displayed significant correlated expression and methylation changes in MS patients. Importantly, silencing of SH3YL1 in primary human CD4 cells demonstrated its influence on T cell activation. Collectively, our strategy based on paired sampling of several cell-types provides a novel approach to increase sensitivity for identifying shared mechanisms altered in CD4 and CD8 cells of relevance in MS in small sized clinical materials.

EnrichedHeatmap: an R/Bioconductor package for comprehensive visualization of genomic signal associations

High-throughput sequencing data are dramatically increasing in volume. Thus, there is urgent need for efficient tools to perform fast and integrative analysis of multiple data types. Enriched heatmap is a specific form of heatmap that visualizes how genomic signals are enriched over specific target regions. It is commonly used and efficient at revealing enrichment patterns especially for high dimensional genomic and epigenomic datasets. We present a new R package named EnrichedHeatmap that efficiently visualizes genomic signal enrichment. It provides advanced solutions for normalizing genomic signals within target regions as well as offering highly customizable visualizations. The major advantage of EnrichedHeatmap is the ability to conveniently generate parallel heatmaps as well as complex annotations, which makes it easy to integrate and visualize comprehensive overviews of the patterns and associations within and between complex datasets. EnrichedHeatmap facilitates comprehensive understanding of high dimensional genomic and epigenomic data. The power of EnrichedHeatmap is demonstrated by visualization of the complex associations between DNA methylation, gene expression and various histone modifications.

Improving stock assessments through data prioritization

The need for “better data” is a common response of stakeholders and managers when confronted with the uncertainty of advice resulting from quantitative stock assessments. Most contemporary stock assessments are based on an integrated analysis of multiple data types, each with their associated cost to collect. Data collection resources are inevitably limited; therefore, it is important to quantify the relative value of increased sampling for alternative data types in terms of improving stock assessments. We approached this universal problem using a simulation study of a hypothetical, amalgam species developed from eight separate stock assessments conducted for species found in southeastern US Atlantic waters. We simulated a population and a stock assessment from the amalgam species and then individually improved alternative data types (indices, age compositions, landings, and discards) by increasing either precision or sample size. We also simulated the effects of increased sampling for alternative groupings of data that might be collected in concert (e.g., commercial, recreational, or survey). Our results show that for the snapper–grouper complex we modeled, age composition data have the largest effect on the accuracy of assessments, with commercial age compositions being the most influential. This is due in part to the relative paucity of age composition data for many southeast US marine stocks, so that modest increases in collection efforts have relatively high benefits for age-based assessment models currently in use for the region. Though this study used data from a particular region of the US, our investigative framework is broadly applicable for quantitatively evaluating the benefits of improved data collection in terms of the precision of stock assessments in any region.

A Systems Biology Approach for Identifying Hepatotoxicant Groups Based on Similarity in Mechanisms of Action and Chemical Structure.

When evaluating compound similarity, addressing multiple sources of information to reach conclusions about common pharmaceutical and/or toxicological mechanisms of action is a crucial strategy. In this chapter, we describe a systems biology approach that incorporates analyses of hepatotoxicant data for 33 compounds from three different sources: a chemical structure similarity analysis based on the 3D Tanimoto coefficient, a chemical structure-based protein target prediction analysis, and a cross-study/cross-platform meta-analysis of in vitro and in vivo human and rat transcriptomics data derived from public resources (i.e., the diXa data warehouse). Hierarchical clustering of the outcome scores of the separate analyses did not result in a satisfactory grouping of compounds considering their known toxic mechanism as described in literature. However, a combined analysis of multiple data types may hypothetically compensate for missing or unreliable information in any of the single data types. We therefore performed an integrated clustering analysis of all three data sets using the R-based tool iClusterPlus. This indeed improved the grouping results. The compound clusters that were formed by means of iClusterPlus represent groups that show similar gene expression while simultaneously integrating a similarity in structure and protein targets, which corresponds much better with the known mechanism of action of these toxicants. Using an integrative systems biology approach may thus overcome the limitations of the separate analyses when grouping liver toxicants sharing a similar mechanism of toxicity.

Abstract 355: The integrated analysis of multiple, high-dimensional data types by joint matrix approximations of rank-1 with applications to liver cancer and glioblastoma

Abstract INTRODUCTION: The aim of this study is to validate a novel algorithm called the Joint Analysis of Many Matrices (JAMM) that detects correlations between data matrices associated with multiple data types acquired from a common set of samples. METHODS: Let A = {A1, A2, …,AN} be a collection of data matrices associated with N different data types acquired from a common set of n samples where each Ai has dimensions pi×n. For each Ai, the JAMM algorithm computes a rank-1 approximation Ai ∼ Ui*X where Ui (pi × 1) is the top eigenarray for Ai and X (1 × n) is the top eigengene common to all the Ai. Since X = ∑transpose(Ui)*X, the rows of each Ai that are most correlated with X (and hence to each other) can be identified by applying the universal threshold Ti = σi* √ [2*log(pi)] to Ui where σi is the standard deviation of Ui. The above procedure is iterated on “residualized” versions of the original data to detect additional signals of cross-correlation until the noise floor is reached. JAMM detection performance was assessed using area under the ROC curve (AUC) on N = 3 simulated data matrices each containing 3 distinct signals that correlated pre-selected rows of each matrix embedded in a Gaussian noise background. Resampling methods and Ingenuity Pathway Analysis were used to assess the significance of signals detected by JAMM in real data sets that juxtaposed gene expression and DNA methylation (liver cancer), mRNA and microRNA expression (glioblastoma), and gene expression and PET imaging features (liver cancer). RESULTS: On simulated data, JAMM consistently outperformed three other algorithms (Generalized Singular Value Decomposition, Canonical Correlation Analysis, and the recently developed JIVE algorithm) based on mean AUC for all 3 signals in each data matrix over 100 simulations. On real data, JAMM analysis of mRNA and microRNA data from 234 glioblastoma samples from The Cancer Genome Atlas recapitulated survival-related tumor subtypes previously discovered based on gene expression alone.Moreover, JAMM analysis of genome-wide expression and DNA methylation data resulted in a compact, methylation-driven, gene expression signature that segregated tumor and normal samples and showed significant enrichment for biological processes related to liver cancer. Finally, JAMM analysis of mRNA and PET imaging data identified a novel liver cancer subtype characterized by dysregulation of pathways involved in bile acid metabolism. CONCLUSIONS: We have demonstrated on simulated and real data sets that the JAMM algorithm for the joint analysis of multiple data types compares favorably with other computational methods in terms of speed, accuracy and biological enrichment. For real data sets, the JAMM algorithm appears to constrain the size of signatures found for each data type with minimal loss of information when compared to signatures obtained from a single data type alone. Citation Format: Gordon S. Okimoto. The integrated analysis of multiple, high-dimensional data types by joint matrix approximations of rank-1 with applications to liver cancer and glioblastoma. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 355. doi:10.1158/1538-7445.AM2014-355

JOINT AND INDIVIDUAL VARIATION EXPLAINED (JIVE) FOR INTEGRATED ANALYSIS OF MULTIPLE DATA TYPES.

Research in several fields now requires the analysis of datasets in which multiple high-dimensional types of data are available for a common set of objects. In particular, The Cancer Genome Atlas (TCGA) includes data from several diverse genomic technologies on the same cancerous tumor samples. In this paper we introduce Joint and Individual Variation Explained (JIVE), a general decomposition of variation for the integrated analysis of such datasets. The decomposition consists of three terms: a low-rank approximation capturing joint variation across data types, low-rank approximations for structured variation individual to each data type, and residual noise. JIVE quantifies the amount of joint variation between data types, reduces the dimensionality of the data, and provides new directions for the visual exploration of joint and individual structure. The proposed method represents an extension of Principal Component Analysis and has clear advantages over popular two-block methods such as Canonical Correlation Analysis and Partial Least Squares. A JIVE analysis of gene expression and miRNA data on Glioblastoma Multiforme tumor samples reveals gene-miRNA associations and provides better characterization of tumor types.

Joint and individual variation explained (JIVE) for integrated analysis of multiple data types

Joint and individual variation explained (JIVE) for integrated analysis of multiple data types