The reading, stated plainly

DNA's readable layer can be read completely and mechanically. One deterministic engine takes a locus and returns its material γ (stacking stiffness), its A4 coordinate, and its R19 switch state, and reproduces bit-for-bit. The reading is closed: every readable channel maps to a measured quantity, and readability ends at a sharp, enumerated boundary — size, sign, dosage, and timing are runtime, never asserted.

This is the whole claim in one place. Complete reading means the admissible reading method for the readable layer is fully specified and closed — not that biology is solved. The boundary where readability ends is part of the reading, not a gap in it; every remaining question is Layer-2 calibration or explicitly out of scope, with no fourth bucket.

What "complete" means for the readable layer, and what it does not

Complete reading means the admissible reading method for the readable layer is fully specified and closed — it does not mean traits are predictable from sequence. The readable layer is exactly four measured things: the material γ (= −mean nearest-neighbour stacking ΔG, SantaLucia 1998), which sets the R19 threshold scale; the A4 coordinate (shell, nearest anchor and its strength, loops, anchor-relative helical phase); the R19 switch state (whether the double-well is bistable — can-fire); and the CpG handles (the writable methylation sites). Read these from any sequence and nothing readable is left over.

Why the boundary is part of the reading

What the sequence does not license is named and never asserted: absolute size, the brake/accelerator sign, dosage, and timing are runtime, set by the environment's tilt h rather than written in the readable material. Knowing exactly where readability ends is itself what it means to have finished the reading. Every question past that edge is either Layer-2 calibration (an absolute scale a measured input would fix) or explicitly out of scope (clinical, phylogenetic, origin-of-life) — there is no fourth bucket.

One operator, five grammar levels — the grammar is mapped to its last level

The same level-and-shape reading is not specific to the cell. Lifted up the tower, one operator — the median-centred, MAD-scaled A4 projection — reads five grammatical levels from sequence, each as a (γ, A4) pair: G1 material (the cell's stacking stiffness), G2 regulatory (which tissue, from transcription-factor syntax — cardiac separates from housekeeping ~14×), G3 architecture (the organ's spatial layout), G4 dynamical (the same locus read by cell state, indexed by CpG-island methylation potential), and G5 body-plan (position on the DNA equals position on the body axis — Hox colinearity, rank correlation 1.000). Read on the time axis, G3 is the developmental order grammar: a gene emerges only after its upstream regulators fire, so emergence order is cascade depth, with the absolute clock pinned by two measured anchors. The full map is Appendix G.

The rest of the paper: the method (Part I), the usable lexicon (Part II), the channels the reading returns including the difference-result (Part III), the proof that the boundary is sharp (Part IV), and the evidence that the reading is real — the same reading grows a body (Part V).