🛠️ Installation Guide
This guide provides step-by-step instructions for setting up GENIUS-LLM. We focus on ensuring a robust installation process across various environments, with a particular emphasis on Linux-based High-Performance Computing (HPC) and Windows (WSL2/Native) systems.
1. Prerequisites
Before starting, please ensure your system meets the following requirements to ensure stability:
- Operating System: Linux (Ubuntu 20.04+, CentOS 7+), macOS, or Windows 10/11.
- Python Version: 3.9, 3.10, or 3.12 (Tested for stability with asynchronous LLM calls).
- Database: MongoDB (v4.4 or higher) must be available as the backend for gene annotation storage.
2. Environment Management (Recommended)
To avoid library version conflicts and ensure reproducibility, we strongly recommend using a Conda virtual environment.
# Create a clean environment
conda create -n genius_llm python=3.12 -y
# Activate the environment
conda activate genius_llm
3. Installation
To install GENIUS-LLM, follow these steps:
# 1. Clone the repository
git clone https://github.com/ZhengJieBioinformatics/GENIUS-LLM.git
cd GENIUS-LLM
# 2. Install dependencies first to avoid failures
pip install -r requirements.txt
# 3. Install GENIUS-LLM in editable mode
pip install -e .
pip install -r requirements.txt
4. Database Setup & Initialization
GENIUS-LLM utilizes MongoDB for high-speed genomic data retrieval. The MongoDB service must be active before running any predictions.
conda install mongodb mongodb-shell mongodb-tools -c conda-forge -y
🔹 For Linux Users (systemd):
Option 1: Users with Root/Sudo Privileges
sudo systemctl start mongod
# Verify status
sudo systemctl status mongod
Option 2: Users without Root Privileges (Local Instance)
# Create a local data directory
mkdir mongodb_data
# Launch MongoDB locally on the default port
mongod --dbpath ~/mongodb_data --port 27017 --fork --logpath ~/mongodb_data/mongodb.log
-
Press Win + R, type services.msc, and hit Enter.
-
Locate MongoDB Server (MongoDB) in the list.
-
Right-click it and select Start.
"Everything set? Move to Quick Start Guide to run your first gene prediction!"