Dataset Information

Automatic Transformation and Integration to Improve Visualization and Discovery of Latent Effects in Imaging Data.

ABSTRACT: Proper data transformation is an essential part of analysis. Choosing appropriate transformations for variables can enhance visualization, improve efficacy of analytical methods, and increase data interpretability. However determining appropriate transformations of variables from high-content imaging data poses new challenges. Imaging data produces hundreds of covariates from each of thousands of images in a corpus. Each of these covariates will have a different distribution and need a potentially different transformation. As such imaging data produces hundreds of covariates, determining an appropriate transformation for each of them is infeasible by hand. In this paper we explore simple, robust, and automatic transformations of high-content image data. A central application of our work is to microenvironment microarray bio-imaging data from the NIH LINCS program. We show that our robust transformations enhance visualization and improve the discovery of substantively relevant latent effects. These transformations enhance analysis of image features individually and also improve data integration approaches when combining together multiple features. We anticipate that the advantages of this work will likely also be realized in the analysis of data from other high-content and highly-multiplexed technologies like Cell Painting or Cyclic Immunofluorescence. Software and further analysis can be found at gjhunt.github.io/rr.

SUBMITTER: Hunt GJ

PROVIDER: S-EPMC8443160 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automatic Transformation and Integration to Improve Visualization and Discovery of Latent Effects in Imaging Data.

Hunt Gregory J GJ Dane Mark A MA Korkola James E JE Heiser Laura M LM Gagnon-Bartsch Johann A JA

Journal of computational and graphical statistics : a joint publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America 20200427 4

Proper data transformation is an essential part of analysis. Choosing appropriate transformations for variables can enhance visualization, improve efficacy of analytical methods, and increase data interpretability. However determining appropriate transformations of variables from high-content imaging data poses new challenges. Imaging data produces hundreds of covariates from each of thousands of images in a corpus. Each of these covariates will have a different distribution and need a potential ...[more]

PMID: 34531645

Similar Datasets

Project description:One of the challenges in the analysis of large data sets, particularly in a population-based setting, is the ability to perform comparisons across projects. This has to be done in such a way that the integrity of each individual project is maintained, while ensuring that the data are comparable across projects. These issues are beginning to be observed in human DNA methylation studies, as the Illumina 450k platform and next generation sequencing-based assays grow in popularity and decrease in price. This increase in productivity is enabling new insights into epigenetics, but also requires the development of pipelines and software capable of handling the large volumes of data. The specific problems inherent in creating a platform for the storage, comparison, integration, and visualization of DNA methylation data include data storage, algorithm efficiency and ability to interpret the results to derive biological meaning from them. Databases provide a ready-made solution to these issues, but as yet no tools exist that that leverage these advantages while providing an intuitive user interface for interpreting results in a genomic context. We have addressed this void by integrating a database to store DNA methylation data with a web interface to query and visualize the database and a set of libraries for more complex analysis. The resulting platform is called DaVIE: Database for the Visualization and Integration of Epigenetics data. DaVIE can use data culled from a variety of sources, and the web interface includes the ability to group samples by sub-type, compare multiple projects and visualize genomic features in relation to sites of interest. We have used DaVIE to identify patterns of DNA methylation in specific projects and across different projects, identify outlier samples, and cross-check differentially methylated CpG sites identified in specific projects across large numbers of samples. A demonstration server has been setup using GEO data at http://echelon.cmmt.ubc.ca/dbaccess/, with login "guest" and password "guest." Groups may download and install their own version of the server following the instructions on the project's wiki.

Dataset Information

Automatic Transformation and Integration to Improve Visualization and Discovery of Latent Effects in Imaging Data.

Publications

Automatic Transformation and Integration to Improve Visualization and Discovery of Latent Effects in Imaging Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets