Networkx graph construction starter
Skill ma-compbio-lab/SkillFoundry/skills/systems-biology/networkx-graph-construction-starter
A framework for discovering, compiling, and validating reusable skills for scientific agents.
npx -y skills add ma-compbio-lab/SkillFoundry --skill networkx-graph-construction-starterAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
1.3 KB, 280 tokens by cl100k_base, as published. Nobody here has run it
NetworkX Graph Construction Starter
Use this skill to build a deterministic toy biological interaction graph and summarize basic network properties with NetworkX.
What it does
- Loads a small TSV edge list representing a pathway-like interaction graph.
- Builds a NetworkX graph with edge attributes.
- Summarizes node and edge counts, connected components, degree centrality, and a shortest path example.
- Writes compact JSON that can feed downstream reporting or agent-planning steps.
When to use it
- You need a local starter for graph-construction workflows in systems biology.
- You want a deterministic network summary before moving on to propagation, enrichment, or causal-network tasks.
Example
python3 skills/systems-biology/networkx-graph-construction-starter/scripts/run_networkx_graph_construction.py \
--input skills/systems-biology/networkx-graph-construction-starter/examples/toy_pathway_edges.tsv \
--source-node EGFR \
--target-node STAT3 \
--out scratch/networkx/graph_summary.json
Verification
- Skill-local tests:
python3 -m unittest discover -s skills/systems-biology/networkx-graph-construction-starter/tests -p 'test_*.py' - Repository smoke:
python3 -m unittest tests.smoke.test_phase25_agent_clinical_proteomics_graph_skills -v