Usage¶

🚀 Quick start¶

▶️ Full pipeline (calculate scores + explore)¶

python -m siflex.main config.json

📊 Calculate scores only¶

python -m siflex.main config.json --explore=none

🌐 Explore network only (skip calculation)¶

python -m siflex.main config.json --no-calculate --explore=network

🧬 Explore pathways only¶

python -m siflex.main config.json --no-calculate --explore=pathways

🔄 Run both explorers simultaneously¶

python -m siflex.main config.json --no-calculate --explore=all

Network explorer: http://localhost:8050
Pathway explorer: http://localhost:8051

⌨️ CLI options¶

python -m siflex.main <config_path> [OPTIONS]

Required:
  config_path              Path to configuration JSON file

Options:
  --no-calculate          Skip score calculation (use existing results)
  --explore=<mode>        Exploration mode: network, pathways, all, or none
                          (default: all)

🖱️ Network explorer¶

The network explorer provides three independent visualisation panels for comparing signalling networks across different conditions:

Tissue Selection: Choose from 26+ cancer tissue types
Cell Line Filtering: Filter by specific cell lines within tissues
Condition Comparison: Compare drug responses across experimental conditions
Pathway Highlighting: Overlay KEGG pathway membership with pie chart visualisations
Interactive Network: Pan, zoom, and hover over nodes to see synergy scores

🧬 Pathway explorer¶

The pathway explorer offers two side-by-side panels for comparing pathway impacts:

Drug Combination Selection: Choose two drugs to analyse
Aggregation Methods: mean, median for node score aggregation
Pathway Ranking: Automatically sort by impact score
Flip Analysis: Identify pathways with opposite responses (synergistic effects)
Interactive Plots: Hover to see detailed pathway information

🐍 Python API¶

from siflex.main import run_siflex, build_config

# Load configuration
config = build_config("config.json")

# Run with custom options
run_siflex(
    config,
    config_path="config.json",
    calculate=True,
    explore="all",
    run_server=True,
)

🧩 Using individual components¶

from siflex.apps import prepare_network_data, run_network_dash

# Prepare network data
nodes, edges, node_pathway_index, PATHWAY_CATALOG, NODE_TO_PATHS, \
    TISSUE_OPTIONS, CELL_LINES_BY_TISSUE, conditions, COND_LABELS, \
    node_scores_agg = prepare_network_data(config_path="config.json")

# Run dashboard on custom port
run_network_dash(
    nodes, edges, node_pathway_index, PATHWAY_CATALOG, NODE_TO_PATHS,
    TISSUE_OPTIONS, CELL_LINES_BY_TISSUE, conditions, COND_LABELS,
    node_scores_agg,
    run_server=True,
    port=8080,
)