https://github.com/danielperez9430/bluegen

Personal genomics from WGS: PRS, ClinVar, pharmacogenomics, ancestry & wellbess reports. Offline, free and powered by PRSKit.
https://github.com/danielperez9430/bluegen

genomic-data-analysis genomics health-report pharmacogenomics polygenic-risk-scores population-genetics prs vcf-automation wgs

Last synced: 29 days ago
JSON representation

Personal genomics from WGS: PRS, ClinVar, pharmacogenomics, ancestry & wellbess reports. Offline, free and powered by PRSKit.

• Host: GitHub
• URL: https://github.com/danielperez9430/bluegen
• Owner: danielperez9430
• License: mit
• Created: 2026-06-08T21:49:13.000Z (about 2 months ago)
• Default Branch: main
• Last Pushed: 2026-06-23T17:24:12.000Z (about 1 month ago)
• Last Synced: 2026-06-23T19:22:21.679Z (about 1 month ago)
• Topics: genomic-data-analysis, genomics, health-report, pharmacogenomics, polygenic-risk-scores, population-genetics, prs, vcf-automation, wgs
• Language: Python
• Homepage:
• Size: 12.1 MB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE

Awesome Lists containing this project

README

          


  



# 🧬 BlueGen

**Personal Genomics Platform** — Powered by the PRSKit polygenic risk score engine.

[![Python](https://img.shields.io/badge/python-3.10+-blue)](https://www.python.org/)

[![License](https://img.shields.io/badge/license-MIT-green)](LICENSE)

[![Version](https://img.shields.io/badge/version-1.1.0-orange)]()

[![PRSKit](https://img.shields.io/badge/engine-PRSKit-purple)]()

Turn a WGS VCF into a comprehensive personal genomics report: polygenic risk scores, pathogenic variants, pharmacogenomics, ancestry, and wellness traits — all offline, all free.

## 🚀 Quick Start

```bash

git clone https://github.com/danielperez9430/BlueGen.git

cd BlueGen

python3 -m venv venv && source venv/bin/activate

pip install -r prs_research_pipeline/requirements.txt

# Full analysis

python3 prs.py run --full --vcf your_sample.vcf.gz

# Interactive dashboard

venv/bin/streamlit run dashboard.py

```

## 📊 What It Does

| Module | Description | Data Source |

|--------|-------------|-------------|

| 🧬 **PRS Engine** | 56 traits, 179 SNPs, population-calibrated z-scores | Curated GWAS + 1000 Genomes |

| 🔬 **ClinVar** | Pathogenic/likely pathogenic variants with confidence tiers | NCBI ClinVar (4.4M records) |

| 💊 **PharmGKB** | Drug-gene interactions, CPIC guideline recommendations | CPIC/DPWG (218 guidelines) |

| 🌍 **Ancestry** | PCA + mtDNA/Y-DNA haplogroups + sub-continental | 1000 Genomes (26 populations) |

| 🦴 **Archaic DNA** | Neanderthal/Denisovan admixture via AADR direct comparison | Allen Ancient DNA Resource (1.23M SNPs) |

| 🩺 **PGS Catalog** | 54 published polygenic scores for complex diseases | PGS Catalog (EBI) |

| 📖 **MedGen** | Disease definitions for ClinVar findings | NCBI MedGen (23K concepts) |

| 📊 **Dashboard** | 6-page interactive Streamlit app | All JSON outputs |

## 📁 Documentation

Full docs: [`prs_research_pipeline/README.md`](prs_research_pipeline/README.md)

Quick reference: [`USAGE.md`](USAGE.md)

Changelog: [`CHANGELOG.md`](CHANGELOG.md)

Data sources: [`prs_research_pipeline/reference/SOURCES.md`](prs_research_pipeline/reference/SOURCES.md)

## 🔧 Requirements

- **Python 3.10+** — `pip install -r prs_research_pipeline/requirements.txt`

- **PLINK v1.90b7.2+** — [Download](https://www.cog-genomics.org/plink/) (free, GPLv3)

- **PLINK 2.0** — [Download](https://www.cog-genomics.org/plink/2.0/) (optional, for advanced QC)

- **bcftools + tabix** — `brew install bcftools tabix` (macOS) or `apt install bcftools tabix` (Linux)

- **BWA** — `brew install bwa` (optional, for FASTQ → BAM alignment)

- macOS/Linux (Windows via WSL2)

- ~200 MB reference data (auto-downloaded on first run)

- ~65 GB reference data bundle (optional — from [archive.org](https://archive.org/details/bluegen-reference-data) snapshot or `prs_research_pipeline/scripts/setup/` to fetch latest from public sources)

  - [`bluegen-reference-data`](https://archive.org/details/bluegen-reference-data) — 1000 Genomes, hg19, ClinVar, MedGen, ClinPGx

  - [`bluegen-pgs-cache`](https://archive.org/details/bluegen-pgs-cache) — 56 PGS Catalog scoring files

  - [`bluegen-archaic-reference`](https://archive.org/details/bluegen-archaic-reference) — AADR 1240K archaic panel (Neanderthal + Denisovan), pre-converted to PLINK

  - [`bluegen-vindija-reference`](https://archive.org/details/bluegen-vindija-reference) — Vindija Neanderthal genome VCFs (chr1–22, hg19, ~44 GB)

  - > **Maintainer:** `python archive_upload.py -j 8` to refresh snapshots

### System Tools

PLINK is **not bundled** — download the correct build for your OS:

| Tool | Version | macOS (Apple Silicon) | macOS (Intel) | Linux |

|------|---------|----------------------|---------------|-------|

| PLINK 1.9 | [v1.90b7.2](https://www.cog-genomics.org/plink/) | `plink` (Rosetta) | `plink_mac` | `plink_linux` |

| PLINK 2.0 | [v2.0.0-a.7.1](https://www.cog-genomics.org/plink/2.0/) | `plink2_mac_arm64` | `plink2_mac` | `plink2_linux` |

Place the binaries in your `PATH` or symlink them into the project root.

## ⚠️ Disclaimer

**RESEARCH USE ONLY — NOT FOR CLINICAL DIAGNOSIS.** This platform is for research and educational purposes. Clinical decisions should not be based on its output without confirmation by a certified laboratory.