GraphMana: graph-native data management for population genomics — benchmark data and pre-built database

Supporting data for the GraphMana manuscript (Mao, 2026). GraphMana is a graph-native data management platform for population genomics that stores variant data as packed genotype arrays in a graph database, enabling incremental sample addition, automatic provenance tracking, cohort management, and export to 17 population genetics formats. This deposit contains: 1kgp_chr22_csv.tar.gz — Pre-built Neo4j-compatible CSV files for 1000 Genomes Project chromosome 22 (1,066,555 variants, 3,202 samples, 5 populations). Import with graphmana load-csv. comprehensive_comprehensive.jsonl — Raw benchmark timing data (JSONL) for all 5 benchmark tasks (incremental addition, cohort extraction, multi-format export, annotation update, lifecycle simulation) comparing GraphMana vs bcftools. bench_comprehensive.py — Python script that generated the benchmark data. validation_results.tar.gz — Downstream tool compatibility validation results (Supplementary Tables 8-9). figure_scripts.tar.gz — R (ggplot2) and Python (matplotlib) scripts for generating all manuscript figures. supplementary.pdf — Complete supplementary materials (15 pages). Software: https://github.com/jfmao/GraphMana PyPI: pip install graphmana