torchdrug
Build TorchDrug Drug Discovery Workflows
Drug discovery teams need reliable choices among molecular datasets, graph models, and evaluation methods. This skill provides focused TorchDrug guidance for complete research workflows.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "torchdrug" from https://skillstore.io/skills/davila7-torchdrug.md and its manifest at https://skillstore.io/api/skills/davila7-torchdrug/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "torchdrug". Choose a workflow for blood-brain barrier prediction.
Expected outcome:
Recommended workflow: use the BBBP dataset, a GIN molecular graph model, scaffold splitting, binary classification, AUROC, and AUPRC.
Using "torchdrug". Select a model for enzyme function prediction from sequences.
Expected outcome:
Start with a pretrained ESM representation and compare it with GearNet when reliable protein structures are available.
Using "torchdrug". My generated molecules are often invalid.
Expected outcome:
Add validity constraints during generation, validate structures with RDKit, filter drug-likeness, and report validity, uniqueness, and novelty separately.
Security Audit
SafeAll 85 static findings were adjudicated as false positives. The alerts represent a PyTorch evaluation-mode call, scientific prose, Markdown formatting, fixed examples, and official resource links.
Risk Factors
⚡ Contains scripts (1)
⚙️ External commands (50)
🌐 Network access (2)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/davila7-torchdrug/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/davila7-torchdrug?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/davila7-torchdrug?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/davila7-torchdrug/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/davila7-torchdrug.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
davila7. (2026). torchdrug security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/davila7-torchdrug/audits/9BibTeX citation
@techreport{davila7-davila7-torchdrug-2026,
author = {davila7},
title = {torchdrug security audit report (audit version 9)},
institution = {Skillstore},
year = {2026},
number = {9},
url = {https://skillstore.io/skills/davila7-torchdrug/audits/9},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "torchdrug security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
- name: "davila7"
date-released: "2026-07-23"
url: "https://skillstore.io/skills/davila7-torchdrug/audits/9"
identifiers:
- type: other
value: "skillstore:davila7-torchdrug:audit:9"
description: "Skillstore immutable audit report identifier"
Compare variants
2 installable variantsEach author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.
Why this variant is first
davila7-torchdrug
2026-08-21
k-dense-ai-torchdrug
2026-08-21
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Screen Molecular Properties
Choose datasets, graph models, splits, and metrics for toxicity, solubility, activity, or blood-brain barrier prediction.
Model Protein Function
Plan sequence-based or structure-based protein workflows using ESM, GearNet, and appropriate protein datasets.
Explore Biomedical Relationships
Design knowledge-graph completion experiments for drug repurposing, disease mechanisms, and gene-disease association analysis.
Try These Prompts
I want to predict {property} from {data_type}. Recommend a TorchDrug dataset, model, task, split, and evaluation metrics.Create a TorchDrug experiment plan for {research_goal}. Include preprocessing, architecture, task configuration, training stages, validation, and expected failure checks.Diagnose this TorchDrug issue: {symptoms}. Review dimensions, graph construction, splitting, metrics, memory use, and model suitability before proposing changes.Design a reproducible TorchDrug pipeline for {objective}. Connect prediction, generation or retrosynthesis, multi-objective ranking, chemistry validation, and uncertainty analysis.Best Practices
- Use scaffold splits for molecular property tasks when realistic generalization matters.
- Match model input dimensions and graph construction to dataset features before training.
- Validate generated molecules and synthesis predictions with chemistry tools and domain experts.
Avoid
- Do not compare models with inconsistent splits, preprocessing, or evaluation metrics.
- Do not treat high benchmark scores as evidence of clinical or experimental validity.
- Do not optimize generated molecules for one property without validity, synthesizability, and safety constraints.
Frequently Asked Questions
What is TorchDrug?
Does this skill install TorchDrug?
Which molecular models are covered?
Can it support protein tasks?
Can predictions replace laboratory validation?
Does it cover knowledge graphs and retrosynthesis?
Developer Details
Author
davila7License
MIT
Skillstore revision
r2
Version notice
The author did not declare a version.
Ref
c43861a65bb95efcae259cd161c9d6f4dc7eec6f
Maintenance freshness
7/25/2026
Usage
6 downloads · 471 views