HEELS: Demo Datasets for Local Heritability Estimation Tutorials

Demo datasets accompany the HEELS software tutorials for estimating local heritability. The data is intended to test two main functions: estimating local heritability and computing low-dimensional representations of linkage disequilibrium matrices. The datasets were released by author Hui Li from Harvard University alongside an open-source Python command line tool.

Use Cases

• Estimating local heritability based on Z-scores from marginal association tests and empirical LD matrices.
• Computing low-dimensional representations of LD matrices for genetic analysis.
• Testing the HEELS software package's functions as described in the accompanying tutorials.

Strengths

• Dataset is explicitly designed for testing the two main functions of the published HEELS software.
• Accompanies documented tutorials available on the project's GitHub wiki pages.
• Associated with a method that attains comparable statistical efficiency to established REML estimators like GCTA and BOLT-REML.

Limitations

• Row count, column names, and file formats are unknown, limiting suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Hui Li, Harvard University

Collection Method

Likely contains summary-level statistics such as Z-scores and empirical linkage disequilibrium data for demonstration.