GenoTools Server

Monorepo containing genomic data processing services and applications for the Global Parkinson's Genetics Program (GP2).

Applications

GP2 Browser

Streamlit web application for browsing GP2 cohort data releases.

• Location: apps/gp2_browser/
• Features: Data releases, quality control, ancestry analysis, SNP metrics, rare variants

Precision Medicine Viewer

Interactive Streamlit application for exploring precision medicine carriers pipeline results.

• Location: apps/precision_med/
• Features: Release overview, genotype viewer, locus reports, probe validation
• Integration: Visualizes output from Precision Med API pipeline

Microservices

GenoTools API

RESTful API interface for genomic data quality control and analysis.

• Location: microservices/genotools_api/
• Features: GenoTools command execution, GCS integration, API key authentication
• Deployment: Docker, Kubernetes (GKE)

GenoTracker

FastAPI + Streamlit application for managing genomic cohort data.

• Location: microservices/genotracker/
• Features: Data visualization, local/GCS data support, interactive exploration

Precision Med API

High-performance genomic carrier screening pipeline for pathogenic variant analysis.

• Location: microservices/precision_med_api/
• Features: Real-time allele harmonization, parallel processing, probe validation, clinical phenotype integration
• Performance: ~10 min extraction for 400 variants across 254 PLINK files
• Data Sources: NBA, WGS, and IMPUTED genotypes with ancestry stratification

Carriers API (DEPRECATED)

API service for genetic carrier analysis.

• Location: microservices/carriers_api/
• Features: Carrier status processing, PLINK integration
• Status: Deprecated in favor of Precision Med API

IDAT Utils

Illumina IDAT file processing toolkit for SNP metrics generation.

• Location: microservices/idat_utils/
• Features: IDAT to VCF conversion, SNP metrics extraction, parquet output

SNP Checksums

Duplicate detection service using SNP-based hashing.

• Location: microservices/snp_checksums/
• Features: Genotype hashing, duplicate identification, PLINK file processing

Infrastructure

• Batch Services: batch_services/
• Deployment: deploy/

Getting Started

Each service contains its own README with specific setup instructions. Services are containerized with Docker and can be deployed to Google Cloud Platform.