skills.rest
jasrajtulsi
jasrajtulsiCOMMUNITY
Explore all skills by this author →
GitHub

torchdrug

Select TorchDrug workflows for graph-based scientific machine learning tasks.

Updated Jul 1, 2026
Search Tags:#graph neural networks#retrosynthesis#protein modeling#knowledge graphs#molecular property prediction#torchdrug
One-click install
npx skills add https://github.com/jasrajtulsi/GRAD-SCOPE --skill torchdrug-jasrajtulsi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: torchdrug
Source: https://github.com/jasrajtulsi/GRAD-SCOPE/tree/main/.claude/skills/torchdrug
Command: npx skills add https://github.com/jasrajtulsi/GRAD-SCOPE --skill torchdrug-jasrajtulsi

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you reason about graph-based machine learning for drug discovery and protein modeling, so you can choose the right datasets, architectures, and training workflows without piecing together scattered documentation.

Core Features & Use Cases

  • Molecular property prediction: Build models for classification and regression tasks on molecular graphs such as BBBP, HIV, Tox21, ESOL, and QM9.
  • Protein modeling: Work with sequence- and structure-based protein tasks including function prediction, stability, localization, fold recognition, and interaction prediction.
  • Knowledge graph reasoning and retrosynthesis: Select embedding models for link prediction, or plan reaction pathways with center identification and synthon completion.
  • Use case: If you need a practical starting point for a new TorchDrug experiment, this Skill helps map your task to the right model family, dataset, and evaluation metrics.

Quick Start

Ask for a TorchDrug setup recommendation for your specific task, dataset, and model choice so you can start the right training workflow immediately.

Frequently Asked Questions about torchdrug

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

FAQPage Schema
How do I choose the right graph neural network architecture for molecular property prediction?▼

Protein modeling applies sequence and structural data to predict function, stability, and fold recognition. It matches protein interaction prediction tasks with suitable model architectures and evaluation metrics for graph-based machine learning workflows.

What is the best way to plan retrosynthesis pathways using graph-based machine learning?▼

Knowledge graph reasoning performs link prediction by selecting appropriate embedding models. You match your knowledge graph datasets with the correct model architectures and evaluation metrics to execute reasoning tasks.

Can I use graph neural networks for both protein modeling and knowledge graph reasoning?▼

Molecular property prediction models require specific dataset matching, such as ESOL or QM9, with appropriate model architectures. You select the correct task interfaces and evaluation metrics to ensure accurate classification and regression training workflows.

How do I set up a TorchDrug experiment for a new drug discovery task?▼

Retrosynthesis workflows utilize center identification and synthon completion to plan reaction pathways. You match your molecular generation requirements with the correct graph-based model architectures to execute and evaluate the synthesis planning.

Related Skills

Explore More Education & Research Skills→
jackspacejackspace
16

torchdrug

Apply graph neural networks to drug discovery tasks with PyTorch.

Community
Advanced
AvaivartikaAvaivartika
1

cheminformatics

Audit molecular ML, QSAR, docking, retrosynthesis, and generation claims in chemistry manuscripts.

Community
Advanced
tassiovaletassiovale
13

torchdrug

Build custom GNN architectures for molecular and protein science tasks.

Community
Advanced
tassiovaletassiovale
13

deepchem

Process molecular data and predict properties using DeepChem.

Community
Advanced