PhenX Measures for Data Sharing, Cross-study Analysis and Data Interoperability

Hamilton, Carol M. (2010) PhenX Measures for Data Sharing, Cross-study Analysis and Data Interoperability. In: Nanoinformatics 2010, November 3 - 5, 2010, Arlington, VA. (Unpublished)

Preview

PDF Carol_Hamilton_TK_Nano_3Nov.v3.pdf - Presentation Download (1MB) | Preview

Abstract

The use of common measures can greatly facilitate data sharing and data interoperability. For genome wide association studies (GWAS), meta-analyses that combine or compare studies is a widely accepted approach. Combining studies can increase the population size and the statistical power of the analysis. Increased statistical power is particularly relevant to the discovery of moderate associations and complex associations, such as gene-gene and gene-environment interactions. Comparing study results from different populations is considered essential to the validation of preliminary findings. The PhenX Toolkit provides the scientific community with high priority, recommended measures for use in GWAS and other large-scale studies. The Toolkit is intended to help researchers expand studies to include (common) measures that are outside of their primary research focus. The PhenX Toolkit currently contains over 200 measures (17 research domains) and will include measures for 21 research domains by the end of 2010. The Toolkit provides users with a web-based interface for searching, browsing, and selecting PhenX measures and protocols. For each PhenX Measure, the Toolkit provides a brief description of the measure, the rationale for selecting the measure, protocol(s) for collecting the measure, and supporting documentation. Measures are associated with a research domain, and may also be included in “Collections” (such as “Top 20 Measures”) and “Conceptual Groups” (such as “Lifestage”) that comprise measures that cut across the research domains. The “Smart Query Tool” offers two options: a high-specificity search through measure and protocol names, synonyms, and keywords; and a high sensitivity full-text search. The “Data Collection Worksheet” enables Toolkit users to easily integrate PhenX measures into their studies and the Data Dictionary provides variable names, identifiers, and attributes in several formats. All PhenX measures have Cancer Biomedical Informatics Grid Common Data Elements (caBIG CDEs) and are accessible using the caBIG CDE browser. Logical Observation Identifiers Names and Codes (LOINC) codes are also being developed for PhenX measures and some have been released with “Trial” status. PhenX measures, protocols and/or variables have been mapped to studies in the database of Genotypes and Phenotypes (dbGaP), the Public Population Project in Genomics’s (P3G) Data Schema and Harmonization Platform for Epidemiological Research (DataSHaPER), and to primary phenotype variables collected by the electronic Medical Records and Genomics (eMERGE) research consortium. The goal is to enable identification of PhenX variables in electroninc medical records (EMRs), clinical data repositories and other resources, promoting data sharing and cross-study analysis. A cross-reference guide linking PhenX measures, protocols and variables to standards and other identifiers is in development. The PhenX Toolkit provides the research community with freely available, well-established measures and bioinformatics tools that promote data sharing and data interoperability.

Actions (login required)

View Item