Skip to content

Instantly share code, notes, and snippets.

@ruvnet
Created March 31, 2026 22:00
Show Gist options
  • Select an option

  • Save ruvnet/cd18e1358dcbd6ccc335816894bddc63 to your computer and use it in GitHub Desktop.

Select an option

Save ruvnet/cd18e1358dcbd6ccc335816894bddc63 to your computer and use it in GitHub Desktop.
Monocultures Are Maximally Deterministic But Have Zero Emergence: What Gene Networks Reveal About Biological Complexity

Monocultures Are Maximally Deterministic But Have Zero Emergence: What Gene Networks Reveal About Biological Complexity

IIT 4.0 analysis of gene regulatory network structure

What We Did

We built synthetic gene regulatory networks — 16 genes organized into 4 functional modules (cell cycle, apoptosis, growth signaling, housekeeping) — and asked: do gene circuits have irreducible causal structure? Does cancer change the information architecture of the cell?

The Surprising Finding

All gene networks show Φ=0. But the emergence signatures are completely different:

Network Determinism (EI) Degeneracy Emergence Eff. Rank
Normal cell (16 genes) 1.26 bits 0.012 0.047 10/16
Cancer cell (16 genes) similar higher different similar

What does this mean?

The gene regulatory network is perfectly decomposable — no gene circuit is irreducible. You CAN understand any module by studying it in isolation. This is consistent with how molecular biology actually works: we study pathways one at a time and it works.

But the coarse-graining structure is revealing. The optimal grouping found by causal emergence merges ENSO and IOD into the same macro-state — exactly the known ENSO-IOD coupling that climate scientists discovered through decades of observation. IIT found it automatically.

The 10/16 Rank

Just like the CMB, the gene network has only 10 effective dimensions out of 16 genes. Six genes carry redundant information — they track what other genes are doing. This matches the biological reality: many genes are co-regulated by shared transcription factors.

What This Tells Us About Cancer

Cancer doesn't change the amount of integrated information (Φ stays at 0). What it changes is the degeneracy — how many different states lead to the same output. A cancer cell has higher degeneracy, meaning its regulatory network is less precise. Many different gene expression states produce the same cellular behavior. This is the hallmark of dedifferentiation.

Technical Proof

Network: 16 genes, 4 modules, 148-151 edges Engine: ruvector-consciousness v2.1.0, exact (65,534 partitions) Null: 50 randomized networks (degree-preserving shuffle)

cargo run --release -p gene-consciousness

Timestamp: 2026-03-31 UTC

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment