技能 string-database
🧬

string-database

安全 🌐 網路存取📁 檔案系統存取

Query STRING protein interaction database

也可從以下取得: K-Dense-AI

Access protein-protein interaction networks from STRING database covering 59M proteins and 20B interactions. Perform functional enrichment analysis, discover interaction partners, and generate network visualizations for systems biology research.

支援: Claude Codex Code(CC)
🥉 74 青銅
1

下載技能 ZIP

2

在 Claude 中上傳

前往 設定 → 功能 → 技能 → 上傳技能

3

開啟並開始使用

測試它

正在使用「string-database」。 Get interaction partners for TP53

預期結果:

  • Top TP53 interactors: MDM2 (score 999), MDM4 (score 995), ATM (score 957)
  • MDM2 is the strongest interactor with experimental evidence

正在使用「string-database」。 Enrich BRCA1 gene list

預期結果:

  • DNA repair (GO:0006281) FDR: 1.2e-15
  • Cell cycle (GO:0007049) FDR: 3.4e-12
  • Homologous recombination (GO:0000724) FDR: 8.9e-10

正在使用「string-database」。 Create network image

預期結果:

  • Generated network image showing TP53, ATM, CHEK2, BRCA1 connections
  • Edges colored by evidence type, thickness by confidence score

安全審計

安全
v5 • 1/17/2026

All static findings are false positives. External command detections are markdown code fences, network access is to legitimate STRING database API, filesystem operations save visualization images, and heuristic alerts are standard API client patterns. No malicious intent detected.

4
已掃描檔案
1,630
分析行數
2
發現項
5
審計總數

風險因素

🌐 網路存取 (1)
📁 檔案系統存取 (1)
審計者: claude 查看審計歷史 →

品質評分

64
架構
90
可維護性
85
內容
30
社群
100
安全
78
規範符合性

你能建構什麼

Protein interaction analysis

Analyze protein networks for gene lists from proteomics or RNA-seq experiments

Pathway enrichment

Identify enriched GO terms and KEGG pathways in protein sets

Interaction partner discovery

Find and visualize proteins interacting with a protein of interest

試試這些提示

Basic network query
Use string-database to get the interaction network for TP53 and MDM2 at high confidence (score 700)
Functional enrichment
Perform GO and KEGG enrichment analysis on these proteins: TP53, BRCA1, ATM, CHEK2, MDM2
Network visualization
Create a confidence-colored network image for these cancer genes and save as png
PPI enrichment
Check if these proteins form a significantly connected module: BRCA1, BRCA2, PALB2, RAD51, ATM

最佳實務

  • Map gene names to STRING IDs first for faster queries
  • Use confidence score threshold appropriate for your analysis goals
  • Wait 1 second between API calls to respect rate limits

避免

  • Do not use low confidence thresholds for publication figures
  • Do not omit species parameter for large protein lists
  • Do not skip error checking when parsing API responses

常見問題

What confidence score should I use?
Use 400 for exploratory analysis, 700 for high-confidence interactions for publications.
Which species are supported?
Over 5000 species including human (9606), mouse (10090), yeast (4932), and E. coli.
How many proteins can I query?
The API handles lists efficiently, but for thousands of proteins use the web interface.
What output formats are available?
TSV, JSON, XML, PSI-MI for data; PNG and SVG for network images.
Is an API key required?
No key needed for basic queries. Registration required for bulk analysis features.
How to cite STRING in publications?
Cite the latest Szklarczyk et al. publication from string-db.org/about