§3 · the coordinate channel
Gamma alone is degenerate: the A4 contact phase is the writable, RNA-targeted channel
γ alone is degenerate. The A4 helical contact phase is about 98% independent of γ (γ tracks GC at R² = 69%, but the contact face only 2%), and real pairs share γ yet differ in contact. RNA is coordinate-targeting: at fixed γ, contact-competence decides whether a fixed drive flips the switch. [V].
Across 12 carriers, γ explains 69% of GC but only 2% of the contact face. A worked pair differs by Δγ = 0.003 yet one coordinate is contact-competent and the other is not — same ruler, different reachability.
Two readings of one locus, only loosely correlated
γ is a thermodynamic average; the A4 coordinate is a structural arrangement. They share some information through GC content, but the part that decides contact — the 3D helical face relative to the anchor — is almost entirely outside γ.
| read | corr. with γ | R² |
|---|---|---|
| γ vs GC content | 0.829 | 69% |
| γ vs CpG O/E | 0.421 | 18% |
| γ vs helical contact face | 0.155 | 2% |
The degeneracy is concrete, not abstract
A real same-γ pair makes the point: with Δγ = 0.003, one member sits at helical face 0.062 and is contact-competent while the other sits at face 0.312 and is not. γ cannot tell them apart; A4 can.
RNA targets a coordinate, not a γ
At one shared γ = 1.3981 (spinodal 0.6363), the same payload 0.5408 flips the contact-competent coordinate (loop-assist 0.1909) ON, yet leaves the non-contact coordinate at the same γ OFF. Contact-competence — not γ — gates reachability.
The environment writes A4; inheritance is of a configuration
γ is sequence-fixed parent→child; the contact state is what toggles ON then OFF, so A4 is the writable channel. At a deep coordinate (γ = 1.5434) a held contact configuration survives one erasure with probability 0.7742 — the inherited object is an A4 coordinate-state, not a rewritten ruler.