Schizophrenia threshold levers — the over-ignition operating-point correction, decomposed into DNA-grounded levers (L1+L3 co-dominant, positive-domain only)

Chapter 24's over-ignition correction is decomposed into a DNA-grounded three-lever target map: modulate inward glutamatergic/NMDA current (L1) and remove the up-stream dopamine drive (L3), co-dominantly, with the outward GABA-A lever (L2) minor, over fourteen schizophrenia genes placed by their promoter switch stiffness. The first L1+L3 co-dominant and first domain-restricted case (positive domain only); targets ranked, never drugs; not medical advice.

The schizophrenia chapter read the disorder's positive symptoms as an over-ignition / aberrant-salience operating point: the ignition fold that should gate a percept or belief sits too low, so weak, internally generated assemblies ignite as hallucinations and delusions. That chapter proved the pole and the corrective sign — the positive domain is an over-ignition pole, so the corrective push is whatever reduces the excess drive or raises the fold, the same direction as epilepsy and bipolar mania — but it left that push abstract, and it carried a second, sharper result the others did not: the disorder splits into three distinct domains (positive / negative / cognitive) that are not a single severity axis. This chapter fills the first gap and respects the second, with the same piece of inherited technology the bipolar, epilepsy and depression levers chapters used: the threshold-shift intervention logic from the analgesic reproducibility package (Zenodo 10.5281/zenodo.20733420). Its three abstract levers — L1 change the inward (excitatory) current, L2 change the outward (potassium / inhibitory) current, L3 change the up-stream drive — carry over unchanged, but schizophrenia is where the frame meets two firsts. It is the first case that is L1+L3 co-dominant: of fourteen schizophrenia genes, six sit on the glutamate/NMDA inward lever (L1) and six on the dopamine up-stream lever (L3), with only two on the channel-restore lever (L2) — the two leading pathophysiologies of the disorder loaded onto the map at once. The L1 set is the glutamatergic axis (GRIN1, GRIN2A, GRIN2B, GRIA3, CACNA1C, CACNB2); the L3 set is the dopamine antipsychotic axis in three sub-axes — receptors (DRD2, DRD4), synthesis/transport (TH, SLC6A3), and serotonergic / catabolic modulation (HTR2A, COMT) — while L2 carries the GABA-A pair (GABRA1, GABRB3). And it is the first case that is domain-restricted: the gain-reducing scalar lever reverses the positive domain only; the negative (output-deficit) and cognitive (long-range wiring) domains are not reached, because a threshold shift cannot re-route geometry nor restore a deficit by lowering a fold. Each gene is placed by reading its own promoter switch stiffness — γ = −mean nearest-neighbour stacking free energy (SantaLucia 1998) over its promoter window, turned into |h_sp| = spinodal(γ) with the frozen engine read-only, and seven of the reads (GRIN2A, CACNA1C, CACNB2 from the bipolar cache, GRIN2B, COMT, HTR2A, GABRA1 from the depression cache) are carried over verbatim. Two honest caveats are recorded, not hidden: the L1 NMDA-hypofunction direction is sign-subtle (the leading model is a parvalbumin-interneuron NMDA hypofunction that disinhibits downstream circuits, so a glycine-site agonist direction appears alongside the naive reduce-excitation sign — the schizophrenia analogue of depression's ketamine caveat), and the HTR2A direction is non-monotone. Three fail-closed disciplines ride along: every dopamine and glutamate link stays graded [O] cited biology, a forbidden-claim scanner rejects any dose / efficacy / safety / synthesis statement (including treats-psychosis, remission, and relapse-prevention language), and a burden-weighted ranking orders targets, never drugs or doses. The firewall is absolute: the promoter |h_sp| is a gene's own switch stiffness, never the §24 network over-ignition threshold, a receptor occupancy, a synaptic dopamine level, a potency, a dose, or a clinical effect. efficacy = 0; not medical advice; the hard problem stays open.

What §24 left abstract

The schizophrenia chapter placed the positive symptoms at the over-ignition operating point of the atlas: health is a population of micro-eddies igniting in gamma only when a percept or belief clears a threshold (the ignition fold), and the positive symptoms of schizophrenia are what happens when that fold sits too low — weak, internally generated assemblies cross it and ignite as aberrant salience: a stray association is experienced as a revelation, an internal voice as an external one. That chapter proved two things and stopped. The first is the pole: the positive domain is an over-ignition pole — the same side of the line as the over-synchronisation disorders — so the corrective sign is whatever reduces the excess drive or raises the fold, the mirror of depression's restore-the-deficit and the twin of epilepsy's and bipolar mania's reduce-the-excess. The second is sharper and is the reason this chapter is unlike the three before it: §24 also proved that schizophrenia is not one axis. Its symptoms split into three distinct domains — positive (the over-ignition just described), negative (an output / gain deficit: blunted affect, avolition, poverty of speech), and cognitive (a long-range wiring / dysconnection problem: working memory, executive control) — and these are not three severities of one thing. The corrective sign §24 proved applies to the positive domain; it says nothing about restoring the negative or re-routing the cognitive. A mechanism atlas should be able to say which molecular targets realise the positive correction and be honest about which domains a threshold lever cannot reach. That is exactly what this chapter does.

The inherited technology, applied a fourth time — and the first L1+L3 co-dominant case

The handle is not invented here and it is not adapted here — it is the same logic the bipolar, epilepsy and depression chapters inherited from the analgesic reproducibility package (Zenodo 10.5281/zenodo.20733420), now turned on a fourth disorder. Its premise is general: an operating point is set by a balance of currents and the drives that bias them, so there are exactly three levers on it. L1 — change the inward, excitatory current. L2 — change the outward, repolarising (or inhibitory) current. L3 — change the up-stream drive that sets where the operating point sits. The frame applies unchanged because the schizophrenia chapter, the three levers chapters, and the analgesic package all share the same R19 substrate — the engine's supercritical pitchfork ṣ = g·s − s³ + h, whose spinodal fold IS the switching barrier.

Schizophrenia makes the frame's generality vivid in a way the first three disorders did not, because of a shift of distribution. Where bipolar leaned on L1 (its replicated GWAS loci are calcium channels), epilepsy on L2 (the KCNQ2/KCNQ3 M-current), and depression on L3 (the up-stream HPA / monoamine / neurotrophic drives, L3-dominant), schizophrenia is the first case that is L1+L3 co-dominant: of the fourteen genes, six sit on L1 and six on L3, with only two on L2. The two leading pathophysiologies of the disorder — the glutamate / NMDA-hypofunction hypothesis and the dopamine hypothesis — load the positive-domain correction onto the map at once, neither one subordinate. That is not a defect of the frame; it is the frame reporting, faithfully, that schizophrenia's positive domain is driven from two directions where the other disorders were driven from one. The pipeline is reused at the level of code, not analogy: seven of the reads — GRIN2A, CACNA1C, CACNB2 from the bipolar cache and GRIN2B, COMT, HTR2A, GABRA1 from the depression cache — are carried over verbatim, because γ is a strand-symmetric property of the sequence and does not change between problems. Nothing about the engine is touched; the module re-emerges the frozen tree read-only and confirms it byte-unchanged, and registers as the eleventh atlas citizen (SZ-T1a-L).

The new finding: a domain-restricted map (the positive domain only)

This is the result that makes schizophrenia the most honest chapter in the levers series, and the model states it as a structural restriction, not a hedge. The three levers are all scalar operators on a single operating point: they raise or lower a current or a drive, and thereby raise or lower the ignition fold. That is precisely the right tool for the positive domain, which §24 defined as a fold set too low — a gain-reducing lever raises the fold and the aberrant ignitions stop crossing it. But it is the wrong tool for the other two domains, and the model says so rather than quietly claiming credit for them. The negative domain is an output / gain deficit: the problem is too little ignition, not too much, so a gain-reducing lever pushes in the wrong direction entirely — you cannot lift a deficit by lowering a fold. The cognitive domain is a long-range wiring problem: the deficit is in the geometry of connectivity (the §19 dysconnection result), and a scalar current or drive lever leaves the locality imbalance exactly invariant — geometry cannot be re-routed by a threshold shift. So the lever map reaches the positive domain and only the positive domain, and the module records this in a domain-restriction witness: positive = reached, negative = not reached, cognitive = not reached. This is the first time the inherited frame has encountered a disorder it can only partly address, and capturing that partiality precisely — axis-structured, not dose-structured — is more informative than a map that pretended to cover all three. It is also why the fail-closed L3-honesty gate for this chapter checks not only that the dopamine links stay [O] but that the co-dominance and the domain restriction are both recorded; it FAILS the build if either slips.

L1 (co-dominant) — the inward glutamatergic / NMDA levers (and the NMDA-hypofunction subtlety)

The first co-dominant lever is the glutamate / NMDA axis, and it carries the chapter's first and most important honest caveat. Six genes sit here. GRIN1 is the obligate NMDA-receptor subunit (GluN1) and the glycine-site node; GRIN2A and GRIN2B are the GluN2A and GluN2B modulatory subunits (GRIN2A is one of the few genes with both genome-wide common-variant and exome rare-variant evidence); GRIA3 is an AMPA-receptor subunit carrying the fast-excitatory sub-route; and CACNA1C (the L-type calcium channel Ca_V1.2, the same top-replicated cross-disorder locus that anchored the bipolar L1) and CACNB2 (its auxiliary β2 subunit) carry the voltage-gated calcium sub-route. The naive reading of L1 would be reduce inward excitatory current — but in schizophrenia the leading glutamatergic model is NMDA hypofunction, and its sign is not that simple. The current understanding is that NMDA receptors on fast-spiking parvalbumin interneurons are hypofunctional, which disinhibits downstream pyramidal circuits and produces the downstream over-ignition the positive symptoms reflect. So the L1 agent direction is sign-subtle: a glycine-site or GRIN1-potentiating agonist direction (raising NMDA function on the interneurons) appears alongside the naive reduce-excitation reading, because the lever's effect runs through an interneuron-then-disinhibition route rather than a direct one. This is the schizophrenia analogue of depression's ketamine caveat, and the model flags it explicitly: the lever placement (the gene is on the inward-current lever) is structural and trait-blind, but the direction of the clinically relevant agent runs through a downstream route and is gene-, dose-, and mechanism-specific, and stays graded [O]. A promoter read places a gene on a lever; it says nothing about whether an agonist or an antagonist is the therapeutic direction. A direction, never a dose.

L3 (co-dominant) — remove the up-stream dopamine drive (the antipsychotic axis)

The second co-dominant lever does not touch a channel; it changes the up-stream dopamine drive that sets the aberrant-salience operating point, and it is the single most established target axis in psychiatry. Six genes sit here, in three sub-axes. The first is the dopamine receptor set: DRD2 — the dopamine D2 receptor, the target shared by every licensed agent for the disorder and a genome-wide significant GWAS locus, the most established single node in the whole map — and DRD4, the D4 receptor whose high clozapine affinity once made it a candidate (a clean D4-selective direction stays exploratory). The second is the dopamine synthesis / transport set: TH (tyrosine hydroxylase, the rate-limiting enzyme — the elevated presynaptic striatal dopamine-synthesis capacity is the most-replicated imaging abnormality in the disorder, and it is a presynaptic handle distinct from the postsynaptic D2 route) and SLC6A3 (the dopamine transporter, DAT). The third is serotonergic and catabolic modulation: HTR2A (the 5-HT2A receptor, the serotonin-dopamine atypical axis, with a non-monotone direction the model records rather than picking a side) and COMT (catechol-O-methyltransferase, the prefrontal catecholamine Val158Met set-point modifier). The mechanism direction across this lever is to reduce the up-stream dopamine drive — lowering the aberrant-salience signal that drops the over-ignition fold — and it reaches the positive domain only, which is exactly why the established route, for all its reach over positive symptoms, leaves the negative and cognitive domains largely untouched. Every L3 link is held at grade [O] — open, cited biology, never derived from the substrate. The model does not claim to compute that dopamine sets the network fold; it records, as cited upstream biology, that this is the drive whose reduction would raise it. A direction, never a dose — efficacy = 0.

L2 — the minor outward GABA-A restore lever

The third lever is minor in schizophrenia — two genes — and it is the restoring mirror of the inward levers, realised through inhibition. GABRA1 (the GABA-A α1 subunit, its read reused from the depression cache verbatim) and GABRB3 (the GABA-A β3 subunit on 15q11–13) raise the inhibitory chloride conductance — the same restoring side of the positive-domain over-ignition axis, approached through the cortical parvalbumin-interneuron deficit that also underlies the L1 sign subtlety. This lever is the direct counterpart of the interneuron-restore idea: where L1 raises NMDA function on the interneurons, L2 raises the inhibitory current those interneurons deliver. It is carried as a minor lever because the replicated genetic and pharmacological weight of the disorder sits on the glutamate and dopamine axes, not on the GABA-A channels — but the two genes are real targets on the restoring side, and they ride L2-adjacent exactly as GABRA1 did in the depression chapter. Every claim on this lever is a mechanism direction only; no opener, modulator, dose, or patient is named.

The DNA grounding: a promoter's own switch stiffness

What places each of the fourteen genes is not a list but a read. For every gene, the module takes its promoter window (transcription start −2000 to +500 bases, Homo sapiens) and computes γ = −mean of the nearest-neighbour base-stacking free energies along that window (the SantaLucia 1998 nearest-neighbour thermodynamics), then turns that γ into the promoter's switch stiffness through the frozen engine's own functions: |h_sp| = spinodal(γ) = 2(γ/3)^1.5 and barrier = γ²/4. The reads span a real range — the stiffest promoter in the set is the dopamine transporter SLC6A3 at γ ≈ 1.60 (|h_sp| ≈ 0.78), followed by the D4 receptor DRD4 (γ ≈ 1.58, |h_sp| ≈ 0.76) and the obligate NMDA subunit GRIN1 (γ ≈ 1.57, |h_sp| ≈ 0.76); the softest is the GABA-A subunit GABRA1 at γ ≈ 1.25 (|h_sp| ≈ 0.54), with the cross-disorder calcium channel CACNA1C also near the soft end (γ ≈ 1.26, |h_sp| ≈ 0.55), and the lead glutamatergic target GRIN2A reading mid-range (γ ≈ 1.47, |h_sp| ≈ 0.69). These are read on the same R19 substrate, with the same engine, that the bipolar, epilepsy, depression and analgesic packages used, which is the whole point of the inheritance: one substrate, one pipeline, now five problems. The γ read is a property of the gene's promoter sequence, blind to whether the gene is on or off and to gain / loss / expression level, and that is all it is.

Ranking targets, and the firewall that keeps gamma honest

The last component prioritises, and it prioritises targets, never drugs or doses. A burden-weighted score combines three declared, cited weights — clinical burden (0.40), unmet need (0.35), and genetic-evidence / druggability (0.25) — on cited 1–5 tiers, and ranks the genes by that score alone. The top actionable target is GRIN2A — the glutamatergic subunit with both common- and rare-variant support, addressing the treatment-resistant and negative/cognitive remainder the dopamine route misses — followed by GRIN2B and then the established dopamine receptor DRD2. A substantive consequence of the unmet-need weight is that it lifts the glutamatergic L1 axis above the established D2 route: the D2 receptor carries the highest burden but the lowest unmet need precisely because it is already the established axis, so the ranking surfaces the NMDA route as the higher-leverage unmet direction — the L1+L3 co-dominant map turned into a priority order, with the unmet-need tier breaking the L1/L3 symmetry in L1's favour. Several genes are flagged not actionable by the generic lever sign: CACNA1C and CACNB2 (cross-disorder calcium set-point), DRD4 (exploratory), GABRB3 (exploratory), and SLC6A3 — the transporter is flagged because DAT blockade raises dopamine (the stimulant direction that worsens the positive domain), the wrong direction entirely. Crucially, the γ read is carried alongside each target as structural context but is never folded into the score — and the result is a clean demonstration of the firewall: the priority ranking and the γ / |h_sp| ranking are decoupled. The stiffest promoter read in the whole set, SLC6A3 (|h_sp| ≈ 0.78), sits near the bottom of priority and is non-actionable; the top-priority target GRIN2A has only a mid-range read (|h_sp| ≈ 0.69). If promoter stiffness drove the ranking, neither could sit where it does. That decoupling is the firewall made visible, and it must be stated once more in full: the promoter |h_sp| is a gene's own switch stiffness, and it is never equated with the §24 network over-ignition threshold on R, nor with a receptor occupancy, a synaptic dopamine level, a compound's potency, a dose, an in-vivo selectivity, or any clinical effect. A fail-closed forbidden-claim scanner guards the whole package: it scans the written results for any dose, efficacy-as-fact, safety-as-fact, or synthesis statement — including treats psychosis, remission, relapse-prevention, and antipsychotic-efficacy language — carries a negation guard, and includes a planted self-test that must fire on its own bait, failing the build if it ever does not. This module reproduces bit-for-bit with the engine byte-unchanged.

Everything here is an in-silico reading of promoter sequence and a frame for organising targets, not a clinical measure, a diagnosis, or a prescription. The model asserts mechanism directions and target placements — an over-ignition fold can be raised two ways at once; these fourteen genes populate the levers; schizophrenia loads them co-dominantly on glutamate (L1) and dopamine (L3); the map reaches the positive domain only; these targets carry the highest genetic and unmet-need burden — and nothing about which agent acts on any lever, at what dose, in whom, whether any real compound changes anyone's symptoms, or that anyone should change a treatment. The agents named as directions (the D2-antagonist route on the receptor sub-axis, the presynaptic-synthesis route, the glycine-site / NMDA-potentiating route flagged sign-subtle on L1, the 5-HT2A route flagged non-monotone) are illustrations of a sign, never a recommendation — and the two recorded caveats (the interneuron-disinhibition route of the L1 NMDA-hypofunction direction, the non-monotone HTR2A direction) are there precisely because a generic lever sign is not a clinical direction. Real schizophrenia is heterogeneous and is three domains, not one — positive, negative, and cognitive, with distinct courses and only partial overlap — and that heterogeneity, including the two domains this lever map explicitly cannot reach, is locked. A promoter read and a lever assignment are mechanism boundaries, not a claim about the felt quality of psychosis (Axis-A firewall — consciousness_claim = 0, the hard problem stays open). This is not medical advice, not a diagnosis, not a treatment protocol, and not a cure. medium_efficacy_tested = 0; targets ranked, never drugs or doses.