Common Cleaning workflows

The following were developed to facilitate analysis of harmonized data files generated from the HHEAR Data Repository. They are R Markdown files, unless otherwise specified. They are not meant for the study investigator-supplied data files or original HHEAR lab files that can also be downloaded from the Repository.

  • Phenotype Cleaning: In the HHEAR Knowledge Graph, outcomes are represented as phenotypes, and harmonized datasets often store multiple phenotype codes for an individual in a single “Phenotype” column, with codes defined in the accompanying codebook. The provided code expands this column into separate indicator variables for each phenotype (1 = present, blank = absent) while retaining the original column; users can combine related indicators (e.g., has vs. does not have asthma) to create binary variables as needed. — View Source R Markdown

Suggested citation text

HHEAR Data Center (2026).«Name of Program (Version No.).» https://github.com/hhear/analytic-resource-library

Example: HHEAR Data Center (2026). Phenotype Cleaning (Version 1.0). https://github.com/hhear/analytic-resource-library