lib-torchdrug

Name: lib-torchdrug
Availability: InStock
Author: biomaps-infra

Develop drug discovery AI models with PyTorch-native graph neural networks.

Updated Mar 13, 2026

Search Tags:#drug discovery #cheminformatics #graph neural networks #molecular modeling #knowledge graphs #protein science

Ready to run

npx skills add https://github.com/biomaps-infra/blender-opencode --skill lib-torchdrug

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: lib-torchdrug
Source: https://github.com/biomaps-infra/blender-opencode/tree/main/.opencode/skills/lib-torchdrug
Command: npx skills add https://github.com/biomaps-infra/blender-opencode --skill lib-torchdrug

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines complex tasks in drug discovery and molecular science by providing a powerful, unified toolkit for building and deploying AI models.

Core Features & Use Cases

Graph Neural Networks: Develop custom GNNs for molecular and protein data.
Task-Specific Modules: Leverage pre-built tasks for property prediction, generation, and knowledge graph reasoning.
Use Case: Predict the binding affinity of novel drug candidates to a target protein using graph neural networks trained on existing binding data.

Quick Start

Use the lib-torchdrug skill to train a GIN model for molecular property prediction on the BBBP dataset.

Frequently Asked Questions about lib-torchdrug

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I train graph neural networks for molecular property prediction?▼

You can train graph neural networks for molecular property prediction using this PyTorch-native skill, which provides pre-built task modules and over 40 curated datasets like BBBP to streamline model development.

Can I use PyTorch and RDKit for drug discovery model development?▼

Yes, this skill requires integration with RDKit for cheminformatics and PyTorch for deep learning, providing a unified toolkit to build AI models for drug discovery and molecular science.

What's the best way to predict binding affinity for novel drug candidates?▼

The best way to predict binding affinity is using graph neural networks trained on existing binding data, a process facilitated by the task-specific property prediction modules in this skill.

Does this support knowledge graph reasoning and protein modeling?▼

Yes, it supports knowledge graph reasoning and protein modeling by providing 20 model architectures and task-specific modules designed for complex bioinformatics and molecular science applications.

How do I build a retrosynthesis planning model using graph neural networks?▼

You can build a retrosynthesis planning model by leveraging the pre-built generation and reasoning tasks included in this skill, utilizing its PyTorch-native graph neural network architectures.

What datasets are available for cheminformatics and molecular generation tasks?▼

Over 40 curated datasets are available for cheminformatics tasks, supporting molecular property prediction, molecular generation, and knowledge graph reasoning out of the box.