string-db

🧬 BioMCP — 生物信息学 MCP 服务器：PubMed/NCBI/BLAST/PDB/UniProt/富集分析，让任意 AI 助手直连生物数据库。Open-source bioinformatics MCP server.

A comprehensive bioinformatics pipeline for downloading and analyzing genes associated with specific phenotypes from multiple biological databases. This package provides both R scripts for gene retrieval and Python analysis tools for downstream analysis.

Extract, cluster and analyse gene expression patterns using GTEx data and STRING-db and Enrichr APIs

Comprehensive bioinformatics analysis of the human PTEN gene and protein using publicly available databases and computational tools.

A GNN-based platform for predicting novel drug-target interactions to accelerate R&D.

MCP server for STRING v12: protein–protein association networks, interaction partners, and functional enrichment analysis — as typed tools for LLM agents.

Protein-network proximity analysis for Parkinson's disease candidate prioritization using STRING and GTEx substantia nigra expression.

Scientific MVP for LUAD target prioritization using knowledge graphs, GraphSAGE, and GraphRAG.

R workflow for patient-specific protein-protein interaction subnetwork profiling from bulk RNA-seq-derived gene tables.

AI-powered drug repurposing pipeline that federates live data across 8 biomedical databases in real time, ranks candidates via DeepSeek V4 LLM analysis, and visualizes target–drug–protein networks in an interactive 3D knowledge graph.

Build a protein-protein interaction network from STRING and rank hub genes by degree centrality (the CytoHubba idea).

Protein evidence workbench — joins UniProt, STRING & Gene Ontology into one annotated report, with in-browser Kyte-Doolittle prediction and an LLM assistant. Single HTML file, no backend.

虚拟小鼠 vmouse · IGFBP4 基因扰动数据包:LINCS/DepMap/IMPC/STRING 等真实公开数据检索+富集(非预测)

Identifying hub and bottleneck genes in pancreatic cancer (GSE16515) via differential expression analysis (GEO2R) and PPI network topology (STRING, Cytoscape, DAVID). Found FN1, CCNB1, CDK1 as hub genes and 11 bottleneck genes as candidate biomarkers for pancreatic ductal adenocarcinoma.

STRING protein-protein interaction networks across ~12k organisms

End‑to‑end pipeline for generating context‑aware drug embeddings from LINCS L1000 using RNA‑seq preprocessing, STRING PPI networks, and GNNs, with UMAP visualisation and drug‑class neighbourhood analysis.

Analyze Parkinson’s disease candidate genes with protein-network proximity using STRING and GTEx substantia nigra data for prioritization

Sequence, domain and interaction-network analysis of SIGMAR1 (ALS16) against a custom BLAST database of 84 amyloid-associated proteins