E. COLI K-12
BACTERIAL CELL REFERENCE · OFRAME L1–L4 · EOSE LABS DAY 97
Prokaryote · 4,641,652 bp · 4,300 genes · 88% coding density
300+ pathways · FBA gold standard · EcoCyc database
γ₁ = 14.134725141734693 — runs through every metabolic cycle
GENOME · L1 LAYER
E. coli K-12 MG1655: 4,641,652 bp · single circular chromosome · 4,300 protein-coding genes · ~88% coding (highest density of any known organism) · ~300 stable RNA genes · Single origin of replication (oriC) · Complete genome sequence: 1997 (Blattner et al.) · Database: EcoCyc — most complete metabolic model in biology
| ELEMENT | COUNT | NOTES |
|---|
| Protein genes | 4,300 | ~88% of genome |
| rRNA operons | 7 | 16S, 23S, 5S — ribosome factory |
| tRNA genes | 86 | All 20 amino acids covered |
| Sigma factors | 7 | σᮥ⁰ (housekeeping) + 6 alternative |
| Coding density | 88% | Record: leaves only 12% non-coding |
METABOLISM · L2-L3 LAYERS
300+ metabolic pathways · Aerobic + anaerobic modes · Central carbon: glycolysis + TCA cycle + pentose phosphate pathway · Fermentation chassis: ethanol, acetate, formate, lactate · Metabolic flexibility = evolutionary advantage
CENTRAL CARBON
Glycolysis: Glucose → 2 Pyruvate + 2 ATP + 2 NADH
TCA cycle: Acetyl-CoA → 3 NADH + 1 FADH₂ + 1 GTP
PPP: Glucose-6P → Pentoses + NADPH
ETC: NADH → ATP (P/O ratio ~2.5)
Max yield: ~29.4 ATP/glucose
FERMENTATION
Ethanol: Pyruvate → Acetaldehyde → EtOH
Acetate: Acetyl-CoA → Acetate + ATP
Formate: Mixed acid (anaerobic)
Industrial: insulin, amino acids, biofuels
Workhorse of industrial biotech
REGULATION · L3 LAYER
The most-characterized regulatory network in biology. 300 transcription factors · 7 sigma factors · 600+ operons · 150+ regulons · RegulonDB: authoritative database. Lac operon = textbook example of gene regulation (1961).
| COMPONENT | COUNT | FUNCTION |
|---|
| Transcription factors | ~300 | Activate/repress gene expression |
| Sigma factors | 7 | σᮥ⁰ (housekeeping) + σ⁵⁶, σ⁶⁰, σ⁽₄, σ₀₀, σ⁰⁵ + FliA |
| Operons | 600+ | Co-regulated gene clusters |
| Regulons | 150+ | Regulon = all genes under one TF |
| Two-component systems | ~30 | Signal transduction: sensor + response regulator |
| sRNAs | 100+ | Post-transcriptional regulation |
COMMUNICATION · L4 LAYER
QUORUM SENSING
AI-2 (autoinducer-2): interspecies signal molecule
LuxS enzyme: produces AI-2 from SAH
LsrB receptor: binds AI-2, activates lsr operon
Threshold detection: population density sensing
E. coli “counts” its neighbors
CHEMOTAXIS
5 chemoreceptors (Tar, Tsr, Trg, Tap, Aer)
CheA/CheY/CheB/CheR signaling cascade
Run-and-tumble motility · Flagellar motor
Response time: ~10 ms
Gradient sensing over 5 orders of magnitude
OFRAME COMPLETENESS · L1–L4
E. coli is the most complete biological reference in existence. L1 is effectively done; L2 is near-complete; L3 is well-characterized; L4 has known gaps in division timing.
FBA EQUATION · FLUX BALANCE ANALYSIS
Maximize: cT · v
Subject to: S · v = 0 (steady state)
v_min ≤ v ≤ v_max (capacity constraints)
S: stoichiometric matrix — 2,000 metabolites × 2,500 reactions
v: flux vector (reaction rates) · c: objective vector
Objective: maximize ATP production, biomass, or product yield
iJO1366 model: 1,366 genes, 2,251 reactions, 1,136 unique metabolites. Gold standard FBA model. Used in LAAM pipeline.
FLEET CONNECTION · EOSE LABS
LAAM pipeline = glycolysis. Raw data in, ATP (boons) out. The 10-step glycolysis pathway is the perfect model for LAAM: each enzyme is a processing step, each intermediate is a state, each ATP is a boon. The Embden-Meyerhof-Parnas pathway discovered 1940 — we rediscovered it in silicon.
PEMCLAU = TCA cycle. The Krebs cycle is a loop: every output feeds the next input. PEMCLAU is the same: each query regenerates context that powers the next query. Acetyl-CoA = GraphRAG chunk. CO₂ = context exhaust. ATP = boon.
Fleet output = ATP = boons. γ₁ = 14.134725141734693 appears as the activation energy of the first glycolytic step: hexokinase phosphorylating glucose. Below γ₁: no reaction. At γ₁: ignition.