BRIGHT

Bacterial Risk Identification and Gene-based Health Toolkit (BRIGHT) is a metagenome-based platform for human pathogenic bacteria (HPB) risk identification. It enables one-click identification of potential HPB, quantitative risk assessment, disease association analysis, and pathogenic gene detection. BRIGHT further supports gene-based identification of bacteria with high human pathogenic risk and integrates machine learning models to associate high-risk bacteria with environmental factors.

Preparation

1. BRIGHT is based on conda environment management. Please ensure the following tools are installed and configured before running the platform:
   • MEGAHIT
   • MetaWRAP
   • dRep
   • CoverM
   • HUMAnN3 (excluding databases)
   • eggNOG
   • GBDB-tk
   • MetaVF_toolkit
   • deepARG
   • mobileOG-db
   • Esymicrobiome
2. The input data must be paired-end metagenomic sequencing reads
3. metadata.txt should be prepared.

Usage

bash /path/to/BRIGHT.sh

Output

BRIDGE provides multi-level human pathogenic bacteria risk outputs, including:

Bacterial and Disease risk file: species_level.tsv
Quantitative risk files: overall_risk.tsv, overall_risk_sample.tsv
Potential pathogenic gene files: VF.category.tsv, VF.type.tsv, KO.PathwayL2.tsv, ARG.type.tsv, mge_ARG_co.tsv and mge_VFG_co.tsv
Envronmental influence: SHAP_feature_importance.csv

Copyright 2025-2026 Qisheng Li, qs.li@cqu.edu.cn, Chongqing University, China; Huan Liu, huanliu@cqu.edu.cn, Chongqing University, China; Meng Liu, liumeng@cqu.edu.cn, Chongqing University, China.