README.md

metadata

license: afl-3.0

🧬 Ubiquitin Carboxy-terminal L1 Hydrolase Target Bioactivity Dataset.

📖 Table of Contents

• ✨ Introduction
• 🔍 Dataset Overview
• 📊 Features
• ⬇️ Downloading the Dataset
  • Direct Download Links
  • Using Git LFS
• 💡 Example
• 🤝 Contributing
• 📜 License
• 📫 Contact

✨ Introduction

Welcome to the Ubiquitin L1 Hydrolase Target Bioactivity Dataset! This open-source dataset is meticulously curated to support machine learning applications in drug discovery and biochemical research focused on Ubiquitin L1 Hydrolase (Ubiquitin L1H). The dataset comprises compounds that have demonstrated activity or inactivity in the Ubiquitin L1 Hydrolase binding site, determined based on their IC₅₀ values. Whether you're developing predictive models or conducting exploratory analysis, this dataset provides high-quality data essential for understanding compound interactions with Ubiquitin L1 Hydrolase.

🔍 Dataset Overview

• Total Samples: 17,421 compounds
  • Active Compounds: 3,272
  • Inactive Compounds: 14,149
• Categorization Criteria:
  • Active: Compounds exhibiting significant binding activity with Ubiquitin L1 Hydrolase based on IC₅₀ ≤ threshold_value μM.
  • Inactive: Compounds showing no significant binding activity with IC₅₀ > threshold_value μM.
• Data Sources:
  • PubChem
  • ChEMBL
  • Custom experimental data
• Features:
  • SMILES: Simplified Molecular-Input Line-Entry System representation
  • Molecular Descriptors: e.g., Molecular Weight, LogP, etc.
  • Bioactivity Scores: IC₅₀ values
  • Target Information: Binding affinity details
• Format: Available in CSV and JSON formats

📊 Features

• Rich Feature Set: Comprehensive molecular descriptors and bioactivity scores for in-depth analysis.
• High-Quality Annotations: Verified and standardized data ensuring reliability.
• Extensive Metadata: Includes references to original data sources and experimental conditions.
• IC₅₀-Based Classification: Clear distinction between active and inactive compounds based on standardized IC₅₀ thresholds.

⬇️ Downloading the Dataset

You can download the dataset directly from Hugging Face or clone the repository to access the data files.

Direct Download Links

Download the dataset files in your preferred format:

• CSV Format:
  • Download active_compounds.csv
  • Download inactive_compounds.csv
• JSON Format:
  • Download active_compounds.json
  • Download inactive_compounds.json

Using Git LFS

If you prefer to clone the repository to access the dataset files, follow these steps:

1. Install Git LFS:

   Git Large File Storage (LFS) is required to handle large dataset files.

   # For macOS
   brew install git-lfs
   
   # For Windows
   choco install git-lfs
   
   # For Linux
   sudo apt-get install git-lfs