Extreme-Tail Precision Does Not Add Information to Gene-Level Association Rankings for Human Height

This reproducibility archive accompanies a numerical audit of gene rankings derived from the 2022 GIANT European height genome-wide association study. It includes scripts for source-text probability parsing and tail-preserving Cauchy combination test calculations, independent validation at 100-decimal-digit precision, comparisons with raw and Bonferroni-adjusted minimum-P rankings and MAGMA outputs, derived gene-level scores, result tables, figures, software environment locks, checksums, and data-acquisition records. The analysis used a fixed universe of 17,108 genes. In this dataset, equal-weight Cauchy rankings closely tracked minimum-P baselines and did not show incremental ranking performance. Provider-controlled source files are not redistributed; official download locations and checksum-verification information are provided.