# HSEF v5.3 review: Super-enhancer Model of Cell Identity

> **Superseded build intermediate (2026-10-14).** This file is a build artifact of the super-enhancer-cell-identity case, not a deliverable. The shipped review is `super-enhancer-cell-identity.review.md` in this same folder, which was adversarially reviewed and approved on 2026-10-14 (see `ADVERSARIAL_REVIEW.super-enhancer-cell-identity.md`). This checkpoint carries a later source-card edition and is not byte-identical to the shipped review; where the two differ, the shipped review governs. Its own adversarial pass is logged in `ADVERSARIAL_REVIEW.current-source-checkpoint.md`. Retained for provenance; do not publish or cite in place of the shipped review.


> Sources: 66 primary reports; claims: 70; verified: 70; unverified: 0. Verification is bounded attribution. Access: 12 partial-body and 54 bounded abstract/status examinations. Literature cutoff: 2026-10-09; source examination UTC date: 2026-10-12. No complete-paper audit or blinded calibration claimed.
> **Adversarial review of this build intermediate (NOT an approval of the review):** Generated by Sol 6.1; this intermediate was adversarially examined by Claude Opus 5.5 (Anthropic) on 2026-10-14 and 7 corrections were applied to it (see ADVERSARIAL REVIEW AND APPROVAL below). That pass was performed on this file in the belief that it was a review, before `current-source-checkpoint.review.md` was established to be a build-tool working filename. Its findings are real and are retained, but the status line in that section does **not** approve the Super-enhancer Model of Cell Identity review. The approved review is `super-enhancer-cell-identity.review.md` in this same folder.

## AT A GLANCE

**Super-enhancer labels identify many important regulatory regions, but native perturbations favor heterogeneous local grammar over a uniquely distinct cell-identity control class.**

| Level | Confidence | Causal role | Momentum | Stability | Divergence |
|---|---|---|---|---|---|
| Β — Associational | Moderate | reinforcing / stabilizing factor | Mechanistic only | Stable | Contested |

**Deciding experiment:** Compare a prespecified multi-lineage panel of high-ChIP clusters and conventional clusters matched for constituent number, occupancy, baseline output, spacing and promoter compatibility using native combinatorial deletions and reconstruction with lineage-state tracking.

**Belief versus evidence:** The registered weakest-link returns Β from SC1 and SC2. SC1 is a reproducible operational association, not evidence for qualitative causal class distinction. SC2 has local mechanistic plausibility at Γ, but one native class-distinction prediction challenge lowers it to Β: called clusters can be dispensable while conventional elements are strongly required, and additive/hierarchical or facilitator rules explain many effects. Native nonadditivity is retained and a universal additive counterclaim is rejected. SC3 is Γ for the prespecified distinct-class disproportionate identity claim, while individual native lineage dependencies have real context-specific causality.

**Best-supported rival:** Facilitator grammar and architectural cluster. Blayney separates classical activators from facilitators through native reconstruction and positional rescue. Moorthy finds strong non-SE regions and inactive called SEs; Shin reveals motif-specific hierarchy and additive tested pairs. These favor heterogeneous internal grammar over either a uniformly additive account or a unique class-wide synergy law.

## COMPETING MECHANISTIC MODELS

**Rank 1: Facilitator grammar and architectural cluster (Ε).** Nonactivating elements and their positions support adjacent activators, with local interaction and recruitment rules. Native alpha-globin reconstruction, positional interchange and heterologous beta-globin facilitator rescue. Facilitators coexist with true local nonadditivity; evidence does not justify asserting that their mechanism excludes synergy.

**Rank 2: Additive or hierarchical locus control region (Ε).** Unequal activators, redundancy, stage dependence and promoter compatibility explain much cluster behavior. Native ESC, Wap, Golli/Mbp, Myc and heart perturbations support heterogeneous local hierarchy. An additive-only universal account fails alpha-globin reconstruction and Fgf5 amplification.

**Rank 3: Transcriptional-load drug sensitivity (Δ).** General coactivator and elongation dependence can make highly active genes drug-sensitive. BET inhibition and degradation separate transcription from contacts; targeted TF loss differs from global CDK loss. These contrasts do not uniquely quantify transcriptional load as the sole determinant; SE contexts can retain selective sensitivity.

**Rank 4: LLPS coactivator condensate (Γ).** Multivalent assemblies concentrate machinery and may enhance selected bursts. Coactivator reconstitution and native Sox2 condensate-distance relationships support plausibility. Liquid-state-specific native rescue is not isolated in examined passages; hubs are rare and broad drugs are pleiotropic.

**Rank 5: Emergent-synergy super-enhancer class (Β).** High-ChIP clusters share a distinct nonadditive mechanism that determines identity beyond conventional clusters. Native local nonadditivity and meaningful selected developmental dependencies are causal positives. Local mechanisms and conventional networks explain much of the behavior without a distinct SE class; one class-distinction falsification applies, not a denial of enhancer dependence.

The tournament does not alter the nominal level. Compatible pathways and flexible alternatives are not automatically proved by explaining observations.

## FULL REVIEW

### Registered claim and scope

Super-enhancers are a functionally distinct regulatory class that disproportionately controls cell-identity genes, rather than a quantitative extension of typical enhancers.

Canonical index weakest-link retained; SC3 adapted from oncogenic necessity to the website cell-identity claim before scoring. Operational calling cannot establish qualitative functional distinction.

Selection: remaining-priority list position 16 from the editorial list of 2026-10-09. Registered claim assessed against original sources. Any carried-forward edition and prior visibility are recorded in the change record; no independent blinded calibration is claimed.

### Evidence and causal interpretation

Rubric fit: Rubric 03 applies to native transcription and lineage function. SC3 is adapted before scoring from the index oncogenic necessity formulation to the website cell-identity claim; cancer fitness, differentiation and terminal output remain distinct. Operational existence is scored as causal evidence, not denial of the measurable category.

The operational definition is useful and reproducible within its chosen calling rules. Whyte maps broad Mediator-rich domains and preferential factor sensitivity; Hnisz extends high-occupancy catalogs across human tissues. Density, length, stitching and nearest-gene assignment remain ingredients of the definition rather than proofs of a discrete causal class. Alternative peak-breadth and AP-1 analyses identify regulatory information that the label can miss. PMID23582322 PMID24119843 PMID32917861 PMID33674350

Hnisz 2015 supplies native constituent deletions, not only maps. Prdm14 E3 loss reduces marks at neighboring constituents, while some constituents have small, large or opposite transcription effects. Reporter combinations and pathway responsiveness show complex local logic; they do not establish class-specific emergent behavior after matching conventional clusters. Native hub editing similarly shows stronger consequences of MYO1D hub loss than nonhub loss, with a smaller CTCF-site effect. PMID25801169 PMID29507293 PMID26766440

Hay 2016 is a meaningful native allelic series, but its single deletions and two chosen double deletions cannot exhaust all interactions. R1 and R2 dominate output, individual knockouts retain adult viability, and combined loss is lethal with very low globin. Domain formation persists. Apparent additivity in that series must be interpreted alongside later reconstruction. PMID27376235 PMID30242161 PMID32075757

Blayney rebuilds the native locus from an enhancerless baseline. R2 alone supplies far less output than predicted from its single-deletion phenotype. Facilitators restore output with position-dependent hierarchy and partial redundancy; moving an activator closer alone does not replace them. Multiple clones and mouse validation support local causality. Crucially, erythroid markers, cell profiles and differentiation remain comparable despite severe globin deficiency. These results establish local nonadditivity without showing an erythroid identity switch or a distinct class beyond conventional multipartite regulation. PMID38101409

Wap demonstrates motif-specific hierarchy. Tested pairs are additive, whereas combined TF-site disruption at the proximal element prevents the remaining cluster from assembling. Later lactation changes enhancer requirements; native promoter compatibility further changes output. Fgf5 and Golli/Mbp supply additional super-additive amplification and largely additive cases. The best account is heterogeneous local interaction rules, rather than a universal additive or universal synergy law. PMID27376239 PMID30285185 PMID32321991 PMID33482114 PMID32790717

Moorthy uses native allele-specific deletions and contrasts strong non-SE regions with called clusters. Effects range from small to nearly complete transcription loss, and one called SE has no effect even when removed with its required neighboring cluster. The intact trans allele limits global-state explanations. Sox2 later separates two major transcription elements from broadly distributed architecture; moderate contact loss preserves expression. These findings expose limits of classification without showing that every enhancer is interchangeable. PMID27895109 PMID35710138 PMID33712417 PMID33334884 PMID36717694

The 2025 alpha-globin inversion tests genuine internal directionality. Single R2 inversion preserves output, whereas a larger cluster-containing inversion redirects activity. Removing the boundary or repositioned Mpg promoter does not restore the original pattern. Optical mapping and local controls support interpretation, but the inversion also repositions genes and elements. Erythroid differentiation remains intact; anemia measures terminal function. A linked correction is retained as an unexamined publication limit. PMID39863595

Native developmental necessity is stronger than expression association alone. The Vsx2 cluster deletion removes bipolar neurons in three independent sublines, supported by single-cell expression, immunostaining, ultrastructure and visual response. Normal gross eye size separates this from progenitor-wide loss. Myc cluster loss disrupts hematopoietic development; selected conserved pluripotency elements and screened enhancers support stem-state dependencies. These are scoped causal positives, even though their class-specific uniqueness and local rescue coverage remain incomplete. PMID31493975 PMID29342133 PMID36161929 PMID33113365 PMID25194570

Cardiac, immune, epidermal, osteoclast and muscle studies expand context but often perturb broad TFs, remodelers or coactivators. Their functional phenotypes do not prove that the high-ChIP label uniquely mediates the effect. The pig muscle abstract reverses standard EDL/soleus descriptors and requires original-text resolution before those labels are reused. Promoter competition, lineage-hidden enhancer roles and stage-specific redundancy must remain visible. PMID33107387 PMID34037670 PMID32857768 PMID37330071 PMID36513810 PMID39013863 PMID40592831 PMID36138028 PMID33462242 PMID32895557 PMID33010223 PMID28575289

Cancer dependencies are a separate comparison. Local amplified MYC and PODXL elements can support malignant growth, while enhancer-directed screening discovers modular fitness elements. Brachyury degradation is more selective than broad CDK inhibition; neuroblastoma TF reprogramming changes state, and Ewing sarcoma predictions fail to identify a reciprocal core circuit. These results support specific targets and local programs, not a universal normal-identity class. PMID26656844 PMID39186674 PMID36351387 PMID30612741 PMID33521702 PMID28650485 PMID34283254 PMID41444391 PMID29258295

Coactivator assemblies and Sox2 condensate proximity support a plausible recruitment mechanism. BET inhibition or BRD4 degradation can strongly affect transcription while most measured enhancer contacts remain. Broad chemical dissolution cannot isolate liquid-state necessity. Multiway SE hubs are rare and promiscuous in recent fixed-cell imaging; this limits dominant hub models without denying local constituent cooperation. Genetic BRD4 deletion can differ sharply from bromodomain inhibition. PMID29930091 PMID38194964 PMID33431820 PMID42607687 PMID23582323 PMID33113340 PMID24445267 PMID23086925 PMID24286025 PMID25677180 PMID34197738

The current revision keeps all 66 primary reports and rereads twelve selected body reports. Overall Β holds under weakest-link aggregation. SC2 now records a class-distinction falsification explicitly and SC3 separates broad class disproportion from local lineage necessity. Historical prose grades and the twelve-body count remain visible only as preserved prior narrative, with the correction log controlling current interpretation.

## HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

### Subclaim matrix and aggregation

| Subclaim | Scope | Level | Decision |
|---|---|---|---|
| SC1 Super-enhancers are operationally distinguishable by reproducible chromatin signal. | Endogenous locus, cell state, temporal resolution and intervention specified before outcomes | Β | High-occupancy clusters are reproducibly measured under fixed calling rules and enrich near cell-identity genes. The category is operational; Β marks the causal association boundary, not uncertainty that the signal is observed. |
| SC2 Super-enhancers functionally differ from conventional enhancers beyond additive or hierarchical effects. | Endogenous locus, cell state, temporal resolution and intervention specified before outcomes | Β | Selected clusters exhibit genuine nonadditivity, facilitators, hierarchy and positional effects. However, equally strong conventional elements and dispensable called clusters challenge a qualitative class distinction beyond those local rules. One index-flagged falsification pattern lowers mechanistic Γ to Β; it does not claim all constituents always add independently. |
| SC3 This distinct regulatory class disproportionately controls cell-identity genes. | Endogenous locus, cell state, temporal resolution and intervention specified before outcomes | Γ | Specific native enhancer-cluster deletions affect lineage formation or function, including loss of retinal bipolar cells after Vsx2-region deletion. These establish narrow causal dependencies. The registered SC3 additionally requires that a distinct SE class disproportionately controls identity; activity-matched conventional-cluster contrasts and class-specific mediation are not established, so that broad transfer remains Γ. |

Preregistered aggregation: weakest-link. The registered weakest-link returns Β from SC1 and SC2. SC1 is a reproducible operational association, not evidence for qualitative causal class distinction. SC2 has local mechanistic plausibility at Γ, but one native class-distinction prediction challenge lowers it to Β: called clusters can be dispensable while conventional elements are strongly required, and additive/hierarchical or facilitator rules explain many effects. Native nonadditivity is retained and a universal additive counterclaim is rejected. SC3 is Γ for the prespecified distinct-class disproportionate identity claim, while individual native lineage dependencies have real context-specific causality. Sensitivities: weakest-link Β; primary-subclaim Β; scope-matched Β.

### Decisive-perturbation inventory and causal tally

[4] Single and two selected double deletions show dominant activators, apparent additivity and retained domain structure. Later reconstruction reveals interactions missed by this limited deletion series; erythroid expression is not identity conversion.

[5] Strong non-SE enhancers and dispensable called SEs coexist; cluster effects and redundancy vary. Selected loci and incomplete combinations do not establish every enhancer's equivalence or a universal additive law.

[6] Tested pairs are additive; combined proximal TF-site loss prevents cluster assembly and Wap induction. Motif and stage specificity matter; high milk-gene output is not a whole-lineage identity switch.

[8] Enhancers require interchangeable position-dependent facilitators for full output; reconstructed alleles rescue function. Erythroid populations retain lineage identity; same-locus platform cannot establish a class-wide distinct mechanism.

[9] Cluster inversion redirects output unlike single R2 inversion; boundary and promoter controls exclude simple insulation. Inversion includes genes and positional changes; no block in erythroid differentiation; linked correction body unexamined.

[10] Two key enhancer regions control output while distributed architecture persists; moderate engineered contact loss preserves RNA. Residual contacts remain; snapshot active-site fraction is not directly measured burst frequency.

[36] Three deletion sublines lack bipolar neurons with functional visual deficits and preserved gross retinal size. Local developmental necessity does not establish class-wide emergent synergy; local rescue not shown in examined passages.

[61] Output changes rapidly despite retained promoter contacts after inhibitor and degrader treatments. Broad chemistry and residual proteins remain; hexanediol is nonspecific and cannot uniquely test liquid state.

Tally: 0 positive qualifying study units across scoped subclaims; 10 screened discriminating contrast units. These are not independent causal lines or proof of full-claim necessity/sufficiency. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Shared cohorts, experimental platforms, transgenic lineages and repeated outcomes do not constitute fully independent causal lines.

No positive field-valid crossing was verified for the registered causal scope in this bounded audit. The proposed registered experiment makes that scope testable. Pivotal native-locus perturbations remain informative; a missing activity-matched comparator or unverified class-specific mediation is an audit limitation, not evidence that enhancer clusters lack regulatory effects or that super-enhancer calling has no descriptive utility.

### Cap audit, precedence and Β/Γ/Δ boundary

Native locus perturbation can establish context-specific causal regulation and lineage consequences without a clinical endpoint. It does not by itself establish a qualitative SE class or disproportionate causal control relative to matched conventional clusters. SC1 descriptive association remains Β; SC2 receives one explicit Γ-to-Β downgrade for the class-distinction prediction challenge. SC3 broad class inference is Γ while retinal and hematopoietic locus-specific causality is preserved separately.

Universal six-cap audit: All six universal caps were considered: cell-only inference, unperturbed association, ectopic expression, biomarker-only intervention, forced transfer/seeding, and inherited-to-sporadic scope transfer. Each binds its offered inference; complementary same-scope causal classes are assessed together rather than re-capped in isolation.

Gamma hard-rule audit: High ChIP intensity and condensate or contact enrichment do not prove a separate functional class. Global factor depletion, drug sensitivity, reporter assays and expression alone do not isolate class-specific native identity causality.

SC1 cap: “Genome-wide association and colocalization without direct perturbation cap at Β.” High-occupancy clusters are reproducibly measured under fixed calling rules and enrich near cell-identity genes. The category is operational; Β marks the causal association boundary, not uncertainty that the signal is observed.

SC3 cap: “Mechanistic plausibility plus in vitro reconstitution or simplified cellular support caps at Γ.” Specific native enhancer-cluster deletions affect lineage formation or function, including loss of retinal bipolar cells after Vsx2-region deletion. These establish narrow causal dependencies. The registered SC3 additionally requires that a distinct SE class disproportionately controls identity; activity-matched conventional-cluster contrasts and class-specific mediation are not established, so that broad transfer remains Γ.

SC1 field-valid definition: “
- Endogenous degron of a condensate component with matched rescue, AND evidence that the specific condensate state (not generic protein loss) drives the phenotype
- Locus-specific disruption of looping with locus-specific effect
- Combinatorial histone-mark perturbation showing direct output dependence
- Critically: specificity against cell-cycle collapse, global transcriptional collapse, and replication stress must be demonstrated” Positive full-scope result: no. High-occupancy clusters are reproducibly measured under fixed calling rules and enrich near cell-identity genes. The category is operational; Β marks the causal association boundary, not uncertainty that the signal is observed.

SC2 field-valid definition: “
- Endogenous degron of a condensate component with matched rescue, AND evidence that the specific condensate state (not generic protein loss) drives the phenotype
- Locus-specific disruption of looping with locus-specific effect
- Combinatorial histone-mark perturbation showing direct output dependence
- Critically: specificity against cell-cycle collapse, global transcriptional collapse, and replication stress must be demonstrated” Positive full-scope result: no. Selected clusters exhibit genuine nonadditivity, facilitators, hierarchy and positional effects. However, equally strong conventional elements and dispensable called clusters challenge a qualitative class distinction beyond those local rules. One index-flagged falsification pattern lowers mechanistic Γ to Β; it does not claim all constituents always add independently.

SC3 field-valid definition: “
- Endogenous degron of a condensate component with matched rescue, AND evidence that the specific condensate state (not generic protein loss) drives the phenotype
- Locus-specific disruption of looping with locus-specific effect
- Combinatorial histone-mark perturbation showing direct output dependence
- Critically: specificity against cell-cycle collapse, global transcriptional collapse, and replication stress must be demonstrated” Positive full-scope result: no. Specific native enhancer-cluster deletions affect lineage formation or function, including loss of retinal bipolar cells after Vsx2-region deletion. These establish narrow causal dependencies. The registered SC3 additionally requires that a distinct SE class disproportionately controls identity; activity-matched conventional-cluster contrasts and class-specific mediation are not established, so that broad transfer remains Γ.

### Upward threshold ladder

SC1 Δ: “Requires at least one endogenous perturbation showing the proposed feature contributes causally to the claimed output.” Satisfied: no. For causal transfer beyond the operational category, prespecify an activity-matched cluster panel and test whether the calling category predicts perturbation response beyond signal, length, constituent number and basal output. Operational measurement itself is already corroborated.

SC2 Δ: “Requires at least one endogenous perturbation showing the proposed feature contributes causally to the claimed output.” Satisfied: no. Combinatorial native loss and reconstruction with matched conventional clusters across independently chosen lineages; demonstrate nonadditive interaction terms uniquely predicted by SE classification after activity, spacing, promoter and developmental-state controls.

SC3 Δ: “Requires at least one endogenous perturbation showing the proposed feature contributes causally to the claimed output.” Satisfied: no. Local cluster loss and rescue with lineage tracing, single-cell state and physiological function in multiple prespecified normal lineages, comparing equally strong conventional clusters and excluding general growth failure.

Upper ladder: Orthogonal endogenous locus edits, native reconstruction, allele controls, dose/calling stratification and physiological lineage outcomes were considered. These can support higher context-specific local regulation, and upper rungs do not require a clinical RCT. The examined evidence lacks class-matched causal prediction and a distinct common mechanism after controlling activity, number, promoter and stage. This is not re-capping an already convergent same-scope case; the observed convergent local contributions have a narrower scope.

Independence: Different assay classes address different errors but do not converge on a unique class-wide mechanism. Related alpha-globin and Wap papers are grouped rather than counted as independent confirmations.

Native reconstruction [8]: Reinsertions and positional swaps distinguish intact-locus loss from direct enhancer and facilitator functions; mouse and ESC-derived erythroid systems remain related.

Native allelic comparator [5]: Unchanged trans allele and strong non-SE elements separate local regulation from global differentiation and the calling label.

Native lineage function [36]: Three deletion sublines, single-cell profiles, immunostaining, electron microscopy and vision separate RNA-only change from a genuine developmental consequence; readouts share the deletion cause.

### Falsification and pitfalls

Alternative mechanisms explain the phenotype more parsimoniously. For the distinct-class SC2 proposition, native cluster equivalence, inactive called SEs and potent non-SE elements (PMID27895109), additive and hierarchical series (PMID27376235, PMID27376239) and facilitator position grammar (PMID38101409) challenge qualitative class specificity. One downgrade Γ to Β is applied. Native facilitator reconstruction and Fgf5 nonadditivity prevent treating a universal additive-only rival as established. Additional downgrade: True.

Historical deletion result treated as exhaustive. Hay tested single deletions and two informative doubles; Blayney adds unseen native configurations.

Hierarchy mistaken for synergy absence. Facilitators and positional grammar can generate nonadditivity; the nominal claim additionally requires class distinction.

Gene expression mistaken for cell identity. Alpha-globin reconstruction and inversion retain erythroid differentiation despite profound output/hematologic effects.

Consensus mismatches and unsupported absence. Broad author-name matches and review citations are rejected; unavailable exact targets do not establish no literature exists.

Anatomical phenotype labels reversed. PMID40592831 abstract assigns conventional EDL/soleus properties in reverse; do not publish those descriptors without original verification.

### Adversarial steelman: above, then below

One rung above, Γ: native facilitator reconstruction, Fgf5 super-additive amplification and retinal lineage deletion clearly exceed mere correlation at selected loci. They motivate a class mechanism, but ordinary enhancer networks also show hierarchy and interdependence, matched class comparisons are missing, and the registered operational SC1 plus class-distinction challenge retain Β for the full conjunction.

One rung below, Α: high-ChIP classification can be circular and broad coactivator drug effects can become a metaphor for causality. However, reproducible cell-state regulatory enrichment, selective native deletions and functional reconstruction provide substantive source support. A single class-distinction falsification is counted once, not repeatedly for every additive or negative paper.

Hold overall Β. Local nonlinear regulation and native lineage dependencies are real; the grade attaches to the distinct-class conjunction, not to whether enhancers regulate identity. Neither universal additivity nor universal emergent synergy explains the observed heterogeneous contexts.

### Final score, causal direction and M/S/D

The registered weakest-link returns Β from SC1 and SC2. SC1 is a reproducible operational association, not evidence for qualitative causal class distinction. SC2 has local mechanistic plausibility at Γ, but one native class-distinction prediction challenge lowers it to Β: called clusters can be dispensable while conventional elements are strongly required, and additive/hierarchical or facilitator rules explain many effects. Native nonadditivity is retained and a universal additive counterclaim is rejected. SC3 is Γ for the prespecified distinct-class disproportionate identity claim, while individual native lineage dependencies have real context-specific causality.

Some cluster features arise during differentiation, but state does not solely produce their function (step 1). Native deletions and reconstruction establish local dependence (step 2). A unique class-wide primary identity switch or rate limit is not established (step 3), nor universal necessity of the class rather than selected elements (step 4). Demonstrated local clusters amplify output and stabilize specified identity programs (step 5).

Momentum: Mechanistic only. Recent reconstruction, inversion and imaging papers are completed reports; no verified named ongoing class-discriminating experiment is asserted.

Stability: Stable. No named pending result is asserted to be likely to move the registered score imminently.

Divergence: Contested. No manufactured field consensus or arithmetic gap. These dimensions do not change level or confidence.

## PROPOSED DECISIVE EXPERIMENTS

Compare a prespecified multi-lineage panel of high-ChIP clusters and conventional clusters matched for constituent number, occupancy, baseline output, spacing and promoter compatibility using native combinatorial deletions and reconstruction with lineage-state tracking.

Readout: Interaction coefficients for nascent output, single-cell lineage state and physiological function across complete sampled allele series. Controls: Native-position and matched-expression rescue, intact trans alleles, equally active conventional clusters, developmentally matched cells, viability and global transcription. Supporting: SE classification predicts reproducible class-specific nonadditive interactions and functional identity dependence after all matched covariates, with specific rescue. Falsifying: Matched conventional clusters show the same interaction grammar, or effects reduce to local promoter, facilitator, stage or load parameters independently of classification.

## KNOWLEDGE GAPS

Test SEs against activity-, length-, constituent-number- and promoter-matched conventional clusters, rather than weak isolated enhancers.

Use reconstruction and higher-order combinations to avoid missing interactions masked by single deletions.

Separate transcriptional output, physiological specialization, lineage formation and maintenance of cell identity.

Define a nonadditive null model on an explicit output scale, with uncertainty, saturation and promoter limits.

Resolve facilitator position, sequence, promoter compatibility and developmental redundancy across independent loci.

Separate coactivator protein dependence from condensate-phase-specific function and drug pleiotropy.

Audit linked correction bodies for the alpha-globin inversion, Myc cluster, Gasperini screen and Du study before publication; resolve the muscle abstract descriptor inconsistency.

## REVIEWER NOTES

66 original primary reports were reread at complete abstract/status depth; twelve have freshly selected original Results paragraph receipts. Copied original XML and source metadata retain their retrieval provenance; copied files alone do not earn fresh read credit. The model-plus-empirical structural study and original genetic-screen method/results report are labeled. Most papers and supplements remain unaudited. Scite quota exhaustion was directly observed in another case during this batch; this case was not retried and has no fresh Scite tally/context. Prior source/grade exposure across the corpus prevents a claim of independent blinded calibration. Revision preserves all prior prose outside the three Step15 scoring blocks and explicitly annotates its historical prose and prior subclaim grades.

Search leads, reviews, protocols and correction notices do not pad the primary count. Downloaded XML does not earn examination credit. Linked notices remain attached; unexamined correction details limit inference. This is a coordinator source/claim consistency assessment, not independent blinded calibration.

## ADVERSARIAL REVIEW AND APPROVAL

> **Scope of this section (lead-adjudicated, 2026-10-14).** This file is a superseded build intermediate, not a
> deliverable. The adversarial pass recorded below was genuinely performed, on this intermediate, before it was
> established that `current-source-checkpoint.review.md` is a build-tool working filename rather than a review.
> The findings and corrections it records are real and are kept as an audit record. The approval line at the end of
> this section applies to this intermediate only and does **not** constitute approval of the Super-enhancer Model of
> Cell Identity review. The approved review is `super-enhancer-cell-identity.review.md` in this same folder, which
> carries its own banner and its own ADVERSARIAL REVIEW AND APPROVAL section from a separate pass
> (`ADVERSARIAL_REVIEW.super-enhancer-cell-identity.md`).

**Authoring model:** Sol 6.1. **Adversarial reviewer:** Claude Opus 5.5 (Anthropic). **Date:** 2026-10-14.

**Filing note.** This file is named `current-source-checkpoint.review.md` and sits inside case folders named
for other slugs. Its content is a review of the **Super-enhancer Model of Cell Identity** (registered claim:
super-enhancers as a functionally distinct regulatory class disproportionately controlling cell-identity
genes), selected at remaining-priority position 16 of the 2026-10-09 editorial list. The generic filename is
a build-tool working name, not the review's subject. No file was moved; the misfiling is reported for the
lead to resolve.

**Scope of checks.** REFERENCE PMIDs verified against PubMed metadata: 66/66 resolve to the stated title,
first author, journal and year; DOIs cross-checked. No reference carries a retraction, retraction-of-
publication or expression-of-concern flag. Load-bearing claims spot-checked against PubMed abstracts and,
where a body receipt exists, against the stored original passage: 10 (the operational super-enhancer
definition and preferential factor sensitivity; the enhancer-cluster deletion series showing 12-92 per cent
effects and that super-enhancer prediction misses most multi-factor elements; the alpha-globin genetic
dissection reporting additive rather than synergistic constituent action; the facilitator reconstruction;
the orientation-dependence inversion; the Wap hierarchy, whose "tested pairs are additive" statement was
confirmed verbatim against the stored Results passage of PMID 27376239; the Golli/Mbp largely additive
result; the Vsx2 bipolar-neuron deletion; BET inhibition separating transcription from enhancer-promoter
contact; and the 2026 single-cell imaging result that multiway super-enhancer hubs are rare and
non-specific). Scoring logic re-derived against `references/hsef-framework.md` and rubric 03: ladder letter
and canonical descriptor, the six universal caps, the pre-registered weakest-link aggregation, the
causal-role derivation and the momentum artifact.

**Findings.**

| ID | Severity | Location | Original -> corrected | Evidence |
|---|---|---|---|---|
| 1 | Major | Decisive-perturbation inventory, closing paragraph after the tally | "Pivotal symptom studies remain informative; missing registration or mediator verification is an audit limitation, not evidence that dopamine has no effects or that antipsychotics lack efficacy." -> "Pivotal native-locus perturbations remain informative; a missing activity-matched comparator or unverified class-specific mediation is an audit limitation, not evidence that enhancer clusters lack regulatory effects or that super-enhancer calling has no descriptive utility." | Template bleed: the original sentence is boilerplate from a dopamine/schizophrenia case and names a drug class and a neurotransmitter that appear nowhere else in this review and have no bearing on its claim. |
| 2 | Minor | Evidence and causal interpretation, rubric-fit line | "Rubric03 applies" -> "Rubric 03 applies" | Builder rendering defect (lost space before a numeral). |
| 3 | Minor | Final score, causal direction and M/S/D, role-derivation paragraph | "(step1)" ... "(step5)" -> "(step 1)" ... "(step 5)" | Same rendering defect, five occurrences. |

**Flagged, not corrected (the fix spans files outside this assignment).**

- **Causal-tally count.** The tally states "10 screened discriminating contrast units", but the inventory
  above it prints eight bracketed items ([4], [5], [6], [8], [9], [10], [36], [61]). The same figure of 10
  appears in `super-enhancer-cell-identity.scorecard.json`
  (`causal_tally_audit.screened_discriminating_contrasts`) and in the finalised
  `super-enhancer-cell-identity.review.md`, which belong to a different review assignment. Either two
  inventory entries were dropped in rendering or the count is wrong; the evidence here does not decide
  which, so no number was changed. The lead should reconcile prose and score card together, across this
  file, the finalised review and its score card.

**No level, confidence, role, momentum, stability or divergence value was changed.** Final score remains
Β — Associational; confidence Moderate; role reinforcing / stabilizing factor.

**Rejected findings (checked, no change).**

1. *Scope-matched sensitivity reported as Β.* First reading: since all three subclaims declare the same
   pre-outcome scope qualifier, "the highest subclaim whose scope matches" would be SC3 (Γ), so the reported
   sensitivity looks wrong. Argued against: the registered claim is conjunctive - a distinct
   class **and** disproportionate control of identity genes - and SKILL.md requires a conjunctive hypothesis
   to be scored on the conjunction, not its easier half. No single subclaim's scope covers that conjunction;
   SC2 and SC3 are its two conjuncts, so the scope-matched value of the conjunction is its weaker conjunct, Β.
   The framework text does not force a change, and the headline is set by the pre-registered weakest-link
   rule in any case. No change.
2. *Weakest-link arithmetic.* min(SC1 Β, SC2 Β, SC3 Γ) = Β; the headline and the derivation wording agree
   with the matrix.
3. *Identical Scope column for SC1-SC3.* The column carries the pre-outcome scope **qualifier**, which is
   legitimately the same for all three subclaims; the differing scopes are stated in the subclaim statements.
   A house-style weakness, not an error.
4. *Momentum.* "Mechanistic only" names its artifact (recent reconstruction, inversion and imaging reports,
   all completed) and correctly declines "Decisive work underway". No change.
5. *Body-examination count.* The review says twelve selected body reports were re-read; twelve
   body-presentation receipts exist and the revision correction log records twelve. An intermediate build
   script carried the figure "ten"; the final log and the receipts agree on twelve, so the review is right.
6. *PMID 40592831 muscle-descriptor reversal.* The review flags that this abstract assigns EDL and soleus
   properties in reverse and refuses to reuse the descriptors. Checked: that is the correct handling of a
   defect in the source, not a defect in the review.
7. *Reference titles at #15 (PMID 33107387) and #24 (PMID 36138028).* PubMed's title field drops the
   italicised gene and marker names (Nppa-Nppb, CD4+CD8+); the review prints the published titles. Rejected
   per the italicised-name convention.

**Residual limitations.** Twelve of 66 reports were examined at partial-body depth and 54 at abstract/status
depth, so claims resting only on unexamined full text (for example the alpha-globin single-knockout
viability detail and the retinal deletion's three sublines) were checked for consistency with the abstract
but could not be re-derived here. This file has no score card of its own; its letters were cross-checked
against the score card of the finalised edition in the same folder, which agrees on Β overall and
Β / Β / Γ for SC1-SC3.

**Approval:** Following adversarial review and the corrections listed above, the revised document is approved. -- Claude Opus 5.5 (Anthropic), 2026-10-14.

## REFERENCES

1. Whyte WA, Orlando DA, Hnisz D et al. Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell. 2013. [PMID 23582322](https://pubmed.ncbi.nlm.nih.gov/23582322/). DOI: 10.1016/j.cell.2013.03.035. Access: full_text_partial. Study: PMID23582322.

2. Hnisz D, Abraham BJ, Lee TI et al. Super-enhancers in the control of cell identity and disease. Cell. 2013. [PMID 24119843](https://pubmed.ncbi.nlm.nih.gov/24119843/). DOI: 10.1016/j.cell.2013.09.053. Access: full_text_partial. Study: PMID24119843.

3. Hnisz D, Schuijers J, Lin CY et al. Convergence of developmental and oncogenic signaling pathways at transcriptional super-enhancers. Molecular cell. 2015. [PMID 25801169](https://pubmed.ncbi.nlm.nih.gov/25801169/). DOI: 10.1016/j.molcel.2015.02.014. Access: full_text_partial. Study: PMID25801169.

4. Hay D, Hughes JR, Babbs C et al. Genetic dissection of the α-globin super-enhancer in vivo. Nature genetics. 2016. [PMID 27376235](https://pubmed.ncbi.nlm.nih.gov/27376235/). DOI: 10.1038/ng.3605. Access: full_text_partial. Study: PMID27376235.

5. Moorthy SD, Davidson S, Shchuka VM et al. Enhancers and super-enhancers have an equivalent regulatory role in embryonic stem cells through regulation of single or multiple genes. Genome research. 2017. [PMID 27895109](https://pubmed.ncbi.nlm.nih.gov/27895109/). DOI: 10.1101/gr.210930.116. Access: full_text_partial. Study: PMID27895109.

6. Shin HY, Willi M, HyunYoo K et al. Hierarchy within the mammary STAT5-driven Wap super-enhancer. Nature genetics. 2016. [PMID 27376239](https://pubmed.ncbi.nlm.nih.gov/27376239/). DOI: 10.1038/ng.3606. Access: full_text_partial. Study: PMID27376239.

7. Huang J, Li K, Cai W et al. Dissecting super-enhancer hierarchy based on chromatin interactions. Nature communications. 2018. [PMID 29507293](https://pubmed.ncbi.nlm.nih.gov/29507293/). DOI: 10.1038/s41467-018-03279-9. Access: full_text_partial. Study: PMID29507293.

8. Blayney JW, Francis H, Rampasekova A et al. Super-enhancers include classical enhancers and facilitators to fully activate gene expression. Cell. 2023. [PMID 38101409](https://pubmed.ncbi.nlm.nih.gov/38101409/). DOI: 10.1016/j.cell.2023.11.030. Access: full_text_partial. Study: PMID38101409.

9. Kassouf MT, Francis HS, Gosden M et al. The α-globin super-enhancer acts in an orientation-dependent manner. Nature communications. 2025. [PMID 39863595](https://pubmed.ncbi.nlm.nih.gov/39863595/). DOI: 10.1038/s41467-025-56380-1. Access: full_text_partial. Study: PMID39863595.

10. Taylor T, Sikorska N, Shchuka VM et al. Transcriptional regulation and chromatin architecture maintenance are decoupled functions at the Sox2 locus. Genes & development. 2022. [PMID 35710138](https://pubmed.ncbi.nlm.nih.gov/35710138/). DOI: 10.1101/gad.349489.122. Access: full_text_partial. Study: PMID35710138.

11. Bahr C, von Paleske L, Uslu VV et al. A Myc enhancer cluster regulates normal and leukaemic haematopoietic stem cell hierarchies. Nature. 2018. [PMID 29342133](https://pubmed.ncbi.nlm.nih.gov/29342133/). DOI: 10.1038/nature25193. Access: abstract_metadata. Study: PMID29342133.

12. Thomas HF, Kotova E, Jayaram S et al. Temporal dissection of an enhancer cluster reveals distinct temporal and functional contributions of individual elements. Molecular cell. 2021. [PMID 33482114](https://pubmed.ncbi.nlm.nih.gov/33482114/). DOI: 10.1016/j.molcel.2020.12.047. Access: abstract_metadata. Study: PMID33482114.

13. Bagheri H, Friedman H, Siminovitch KA et al. Transcriptional regulators of the Golli/myelin basic protein locus integrate additive and stealth activities. PLoS genetics. 2020. [PMID 32790717](https://pubmed.ncbi.nlm.nih.gov/32790717/). DOI: 10.1371/journal.pgen.1008752. Access: abstract_metadata. Study: PMID32790717.

14. Zeng X, Lee HK, Wang C et al. The interdependence of mammary-specific super-enhancers and their native promoters facilitates gene activation during pregnancy. Experimental & molecular medicine. 2020. [PMID 32321991](https://pubmed.ncbi.nlm.nih.gov/32321991/). DOI: 10.1038/s12276-020-0425-x. Access: abstract_metadata. Study: PMID32321991.

15. Man JCK, van Duijvenboden K, Krijger PHL et al. Genetic Dissection of a Super Enhancer Controlling the Nppa-Nppb Cluster in the Heart. Circulation research. 2021. [PMID 33107387](https://pubmed.ncbi.nlm.nih.gov/33107387/). DOI: 10.1161/CIRCRESAHA.120.317045. Access: abstract_metadata. Study: PMID33107387.

16. Wang HF, Warrier T, Farran CA et al. Defining Essential Enhancers for Pluripotent Stem Cells Using a Features-Oriented CRISPR-Cas9 Screen. Cell reports. 2020. [PMID 33113365](https://pubmed.ncbi.nlm.nih.gov/33113365/). DOI: 10.1016/j.celrep.2020.108309. Access: abstract_metadata. Study: PMID33113365.

17. Agrawal P, Blinka S, Pulakanti K et al. Genome editing demonstrates that the -5 kb Nanog enhancer regulates Nanog expression by modulating RNAPII initiation and/or recruitment. The Journal of biological chemistry. 2021. [PMID 33334884](https://pubmed.ncbi.nlm.nih.gov/33334884/). DOI: 10.1074/jbc.RA120.015152. Access: abstract_metadata. Study: PMID33334884.

18. Zhang J, Zhou Y, Yue W et al. Super-enhancers conserved within placental mammals maintain stem cell pluripotency. Proceedings of the National Academy of Sciences of the United States of America. 2022. [PMID 36161929](https://pubmed.ncbi.nlm.nih.gov/36161929/). DOI: 10.1073/pnas.2204716119. Access: abstract_metadata. Study: PMID36161929.

19. Singh G, Mullany S, Moorthy SD et al. A flexible repertoire of transcription factor binding sites and a diversity threshold determines enhancer activity in embryonic stem cells. Genome research. 2021. [PMID 33712417](https://pubmed.ncbi.nlm.nih.gov/33712417/). DOI: 10.1101/gr.272468.120. Access: abstract_metadata. Study: PMID33712417.

20. Chakraborty S, Kopitchinski N, Zuo Z et al. Enhancer-promoter interactions can bypass CTCF-mediated boundaries and contribute to phenotypic robustness. Nature genetics. 2023. [PMID 36717694](https://pubmed.ncbi.nlm.nih.gov/36717694/). DOI: 10.1038/s41588-022-01295-6. Access: abstract_metadata. Study: PMID36717694.

21. Proudhon C, Snetkova V, Raviram R et al. Active and Inactive Enhancers Cooperate to Exert Localized and Long-Range Control of Gene Regulation. Cell reports. 2016. [PMID 27239026](https://pubmed.ncbi.nlm.nih.gov/27239026/). DOI: 10.1016/j.celrep.2016.04.087. Access: abstract_metadata. Study: PMID27239026.

22. Lee HK, Willi M, Shin HY et al. Progressing super-enhancer landscape during mammary differentiation controls tissue-specific gene regulation. Nucleic acids research. 2018. [PMID 30285185](https://pubmed.ncbi.nlm.nih.gov/30285185/). DOI: 10.1093/nar/gky891. Access: abstract_metadata. Study: PMID30285185.

23. Metser G, Shin HY, Wang C et al. An autoregulatory enhancer controls mammary-specific STAT5 functions. Nucleic acids research. 2016. [PMID 26446995](https://pubmed.ncbi.nlm.nih.gov/26446995/). DOI: 10.1093/nar/gkv999. Access: abstract_metadata. Study: PMID26446995.

24. Feng D, Chen Y, Dai R et al. Chromatin organizer SATB1 controls the cell identity of CD4+ CD8+ double-positive thymocytes by regulating the activity of super-enhancers. Nature communications. 2022. [PMID 36138028](https://pubmed.ncbi.nlm.nih.gov/36138028/). DOI: 10.1038/s41467-022-33333-6. Access: abstract_metadata. Study: PMID36138028.

25. Li M, Huang H, Li L et al. Core transcription regulatory circuitry orchestrates corneal epithelial homeostasis. Nature communications. 2021. [PMID 33462242](https://pubmed.ncbi.nlm.nih.gov/33462242/). DOI: 10.1038/s41467-020-20713-z. Access: abstract_metadata. Study: PMID33462242.

26. Huang J, Liu X, Li D et al. Dynamic Control of Enhancer Repertoires Drives Lineage and Stage-Specific Transcription during Hematopoiesis. Developmental cell. 2016. [PMID 26766440](https://pubmed.ncbi.nlm.nih.gov/26766440/). DOI: 10.1016/j.devcel.2015.12.014. Access: abstract_metadata. Study: PMID26766440.

27. Peng XL, So KK, He L et al. MyoD- and FoxO3-mediated hotspot interaction orchestrates super-enhancer activity during myogenic differentiation. Nucleic acids research. 2017. [PMID 28575289](https://pubmed.ncbi.nlm.nih.gov/28575289/). DOI: 10.1093/nar/gkx488. Access: abstract_metadata. Study: PMID28575289.

28. Maezawa S, Sakashita A, Yukawa M et al. Super-enhancer switching drives a burst in gene expression at the mitosis-to-meiosis transition. Nature structural & molecular biology. 2020. [PMID 32895557](https://pubmed.ncbi.nlm.nih.gov/32895557/). DOI: 10.1038/s41594-020-0488-3. Access: abstract_metadata. Study: PMID32895557.

29. Wuelling M, Neu C, Thiesen AM et al. Epigenetic Mechanisms Mediating Cell State Transitions in Chondrocytes. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research. 2021. [PMID 33534175](https://pubmed.ncbi.nlm.nih.gov/33534175/). DOI: 10.1002/jbmr.4263. Access: abstract_metadata. Study: PMID33534175.

30. Milner JJ, Toma C, Quon S et al. Bromodomain protein BRD4 directs and sustains CD8 T cell differentiation during infection. The Journal of experimental medicine. 2021. [PMID 34037670](https://pubmed.ncbi.nlm.nih.gov/34037670/). DOI: 10.1084/jem.20202512. Access: abstract_metadata. Study: PMID34037670.

31. Oda Y, Nguyen T, Hata A et al. Deletion of Mediator 1 suppresses TGFβ signaling leading to changes in epidermal lineages and regeneration. PloS one. 2020. [PMID 32857768](https://pubmed.ncbi.nlm.nih.gov/32857768/). DOI: 10.1371/journal.pone.0238076. Access: abstract_metadata. Study: PMID32857768.

32. Oda Y, Wong CT, Oh DH et al. Vitamin D receptor cross-talk with p63 signaling promotes epidermal cell fate. The Journal of steroid biochemistry and molecular biology. 2023. [PMID 37330071](https://pubmed.ncbi.nlm.nih.gov/37330071/). DOI: 10.1016/j.jsbmb.2023.106352. Access: abstract_metadata. Study: PMID37330071.

33. Bae S, Kim K, Kang K et al. RANKL-responsive epigenetic mechanism reprograms macrophages into bone-resorbing osteoclasts. Cellular & molecular immunology. 2023. [PMID 36513810](https://pubmed.ncbi.nlm.nih.gov/36513810/). DOI: 10.1038/s41423-022-00959-x. Access: abstract_metadata. Study: PMID36513810.

34. Du J, Liu Y, Sun J et al. ARID1A safeguards the canalization of the cell fate decision during osteoclastogenesis. Nature communications. 2024. [PMID 39013863](https://pubmed.ncbi.nlm.nih.gov/39013863/). DOI: 10.1038/s41467-024-50225-z. Access: abstract_metadata. Study: PMID39013863.

35. Tan B, Hong L, Xiao L et al. Rewiring of 3D chromatin topology orchestrates transcriptional reprogramming in muscle fiber-type specification and transformation. Nature communications. 2025. [PMID 40592831](https://pubmed.ncbi.nlm.nih.gov/40592831/). DOI: 10.1038/s41467-025-60866-3. Access: abstract_metadata. Study: PMID40592831.

36. Norrie JL, Lupo MS, Xu B et al. Nucleome Dynamics during Retinal Development. Neuron. 2019. [PMID 31493975](https://pubmed.ncbi.nlm.nih.gov/31493975/). DOI: 10.1016/j.neuron.2019.08.002. Access: full_text_partial. Study: PMID31493975.

37. Sciumè G, Mikami Y, Jankovic D et al. Rapid Enhancer Remodeling and Transcription Factor Repurposing Enable High Magnitude Gene Induction upon Acute Activation of NK Cells. Immunity. 2020. [PMID 33010223](https://pubmed.ncbi.nlm.nih.gov/33010223/). DOI: 10.1016/j.immuni.2020.09.008. Access: abstract_metadata. Study: PMID33010223.

38. van Groningen T, Koster J, Valentijn LJ et al. Neuroblastoma is composed of two super-enhancer-associated differentiation states. Nature genetics. 2017. [PMID 28650485](https://pubmed.ncbi.nlm.nih.gov/28650485/). DOI: 10.1038/ng.3899. Access: abstract_metadata. Study: PMID28650485.

39. Kochat V, Raman AT, Landers SM et al. Enhancer reprogramming in PRC2-deficient malignant peripheral nerve sheath tumors induces a targetable de-differentiated state. Acta neuropathologica. 2021. [PMID 34283254](https://pubmed.ncbi.nlm.nih.gov/34283254/). DOI: 10.1007/s00401-021-02341-z. Access: abstract_metadata. Study: PMID34283254.

40. Sheppard HE, Dall'Agnese A, Park WD et al. Targeted brachyury degradation disrupts a highly specific autoregulatory program controlling chordoma cell identity. Cell reports. Medicine. 2021. [PMID 33521702](https://pubmed.ncbi.nlm.nih.gov/33521702/). DOI: 10.1016/j.xcrm.2020.100188. Access: abstract_metadata. Study: PMID33521702.

41. Herranz D, Ambesi-Impiombato A, Palomero T et al. A NOTCH1-driven MYC enhancer promotes T cell development, transformation and acute lymphoblastic leukemia. Nature medicine. 2014. [PMID 25194570](https://pubmed.ncbi.nlm.nih.gov/25194570/). DOI: 10.1038/nm.3665. Access: abstract_metadata. Study: PMID25194570.

42. Grossetete S, Zaidi S, Orth MF et al. The hegemonic EWSR1::ETS oncoprotein overrules core regulatory circuitry principles in Ewing sarcoma. NPJ precision oncology. 2025. [PMID 41444391](https://pubmed.ncbi.nlm.nih.gov/41444391/). DOI: 10.1038/s41698-025-01226-8. Access: abstract_metadata. Study: PMID41444391.

43. Mack SC, Pajtler KW, Chavez L et al. Therapeutic targeting of ependymoma as informed by oncogenic enhancer profiling. Nature. 2018. [PMID 29258295](https://pubmed.ncbi.nlm.nih.gov/29258295/). DOI: 10.1038/nature25169. Access: abstract_metadata. Study: PMID29258295.

44. Zhang X, Choi PS, Francis JM et al. Identification of focally amplified lineage-specific super-enhancers in human epithelial cancers. Nature genetics. 2016. [PMID 26656844](https://pubmed.ncbi.nlm.nih.gov/26656844/). DOI: 10.1038/ng.3470. Access: abstract_metadata. Study: PMID26656844.

45. Lewis MW, King CM, Wisniewska K et al. CRISPR Screening of Transcribed Super-Enhancers Identifies Drivers of Triple-Negative Breast Cancer Progression. Cancer research. 2024. [PMID 39186674](https://pubmed.ncbi.nlm.nih.gov/39186674/). DOI: 10.1158/0008-5472.CAN-23-3995. Access: abstract_metadata. Study: PMID39186674.

46. Chen PB, Fiaux PC, Zhang K et al. Systematic discovery and functional dissection of enhancers needed for cancer cell fitness and proliferation. Cell reports. 2022. [PMID 36351387](https://pubmed.ncbi.nlm.nih.gov/36351387/). DOI: 10.1016/j.celrep.2022.111630. Access: abstract_metadata. Study: PMID36351387.

47. Gasperini M, Hill AJ, McFaline-Figueroa JL et al. A Genome-wide Framework for Mapping Gene Regulation via Cellular Genetic Screens. Cell. 2019. [PMID 30612741](https://pubmed.ncbi.nlm.nih.gov/30612741/). DOI: 10.1016/j.cell.2018.11.029. Access: abstract_metadata. Study: PMID30612741.

48. Seo J, Koçak DD, Bartelt LC et al. AP-1 subunits converge promiscuously at enhancers to potentiate transcription. Genome research. 2021. [PMID 33674350](https://pubmed.ncbi.nlm.nih.gov/33674350/). DOI: 10.1101/gr.267898.120. Access: abstract_metadata. Study: PMID33674350.

49. Saravanan B, Soota D, Islam Z et al. Ligand dependent gene regulation by transient ERα clustered enhancers. PLoS genetics. 2020. [PMID 31905229](https://pubmed.ncbi.nlm.nih.gov/31905229/). DOI: 10.1371/journal.pgen.1008516. Access: abstract_metadata. Study: PMID31905229.

50. Fox S, Myers JA, Davidson C et al. Hyperacetylated chromatin domains mark cell type-specific genes and suggest distinct modes of enhancer function. Nature communications. 2020. [PMID 32917861](https://pubmed.ncbi.nlm.nih.gov/32917861/). DOI: 10.1038/s41467-020-18303-0. Access: abstract_metadata. Study: PMID32917861.

51. Dowen JM, Bilodeau S, Orlando DA et al. Multiple structural maintenance of chromosome complexes at transcriptional regulatory elements. Stem cell reports. 2013. [PMID 24286025](https://pubmed.ncbi.nlm.nih.gov/24286025/). DOI: 10.1016/j.stemcr.2013.09.002. Access: abstract_metadata. Study: PMID24286025.

52. Ing-Simmons E, Seitan VC, Faure AJ et al. Spatial enhancer clustering and regulation of enhancer-proximal genes by cohesin. Genome research. 2015. [PMID 25677180](https://pubmed.ncbi.nlm.nih.gov/25677180/). DOI: 10.1101/gr.184986.114. Access: abstract_metadata. Study: PMID25677180.

53. Brown JM, Roberts NA, Graham B et al. A tissue-specific self-interacting chromatin domain forms independently of enhancer-promoter interactions. Nature communications. 2018. [PMID 30242161](https://pubmed.ncbi.nlm.nih.gov/30242161/). DOI: 10.1038/s41467-018-06248-4. Access: abstract_metadata. Study: PMID30242161.

54. Chiariello AM, Bianco S, Oudelaar AM et al. A Dynamic Folded Hairpin Conformation Is Associated with α-Globin Activation in Erythroid Cells. Cell reports. 2020. [PMID 32075757](https://pubmed.ncbi.nlm.nih.gov/32075757/). DOI: 10.1016/j.celrep.2020.01.044. Access: abstract_metadata. Study: PMID32075757.

55. Vos ESM, Valdes-Quezada C, Huang Y et al. Interplay between CTCF boundaries and a super enhancer controls cohesin extrusion trajectories and gene expression. Molecular cell. 2021. [PMID 34197738](https://pubmed.ncbi.nlm.nih.gov/34197738/). DOI: 10.1016/j.molcel.2021.06.008. Access: abstract_metadata. Study: PMID34197738.

56. Sabari BR, Dall'Agnese A, Boija A et al. Coactivator condensation at super-enhancers links phase separation and gene control. Science (New York, N.Y.). 2018. [PMID 29930091](https://pubmed.ncbi.nlm.nih.gov/29930091/). DOI: 10.1126/science.aar3958. Access: abstract_metadata. Study: PMID29930091.

57. Du M, Stitzinger SH, Spille JH et al. Direct observation of a condensate effect on super-enhancer controlled gene bursting. Cell. 2024. [PMID 38194964](https://pubmed.ncbi.nlm.nih.gov/38194964/). DOI: 10.1016/j.cell.2023.12.005. Access: abstract_metadata. Study: PMID38194964.

58. Le DJ, Hafner A, Raja Venkatesh A et al. Single-cell 3D genome imaging shows super-enhancer hubs are rare, nonspecific, and neither necessary nor sufficient for transcriptional bursting. Cell genomics. 2026. [PMID 42607687](https://pubmed.ncbi.nlm.nih.gov/42607687/). DOI: 10.1016/j.xgen.2026.101331. Access: abstract_metadata. Study: PMID42607687.

59. Kai Y, Li BE, Zhu M et al. Mapping the evolving landscape of super-enhancers during cell differentiation. Genome biology. 2021. [PMID 34526084](https://pubmed.ncbi.nlm.nih.gov/34526084/). DOI: 10.1186/s13059-021-02485-x. Access: abstract_metadata. Study: PMID34526084.

60. Lovén J, Hoke HA, Lin CY et al. Selective inhibition of tumor oncogenes by disruption of super-enhancers. Cell. 2013. [PMID 23582323](https://pubmed.ncbi.nlm.nih.gov/23582323/). DOI: 10.1016/j.cell.2013.03.036. Access: abstract_metadata. Study: PMID23582323.

61. Crump NT, Ballabio E, Godfrey L et al. BET inhibition disrupts transcription but retains enhancer-promoter contact. Nature communications. 2021. [PMID 33431820](https://pubmed.ncbi.nlm.nih.gov/33431820/). DOI: 10.1038/s41467-020-20400-z. Access: full_text_partial. Study: PMID33431820.

62. Rodriguez RM, Suarez-Alvarez B, Salvanés R et al. Role of BRD4 in hematopoietic differentiation of embryonic stem cells. Epigenetics. 2014. [PMID 24445267](https://pubmed.ncbi.nlm.nih.gov/24445267/). DOI: 10.4161/epi.27711. Access: abstract_metadata. Study: PMID24445267.

63. Kim SY, Zhang X, Schiattarella GG et al. Epigenetic Reader BRD4 (Bromodomain-Containing Protein 4) Governs Nucleus-Encoded Mitochondrial Transcriptome to Regulate Cardiac Function. Circulation. 2020. [PMID 33113340](https://pubmed.ncbi.nlm.nih.gov/33113340/). DOI: 10.1161/CIRCULATIONAHA.120.047239. Access: abstract_metadata. Study: PMID33113340.

64. Zhang W, Prakash C, Sum C et al. Bromodomain-containing protein 4 (BRD4) regulates RNA polymerase II serine 2 phosphorylation in human CD4+ T cells. The Journal of biological chemistry. 2012. [PMID 23086925](https://pubmed.ncbi.nlm.nih.gov/23086925/). DOI: 10.1074/jbc.M112.413047. Access: abstract_metadata. Study: PMID23086925.

65. Farooq U, Saravanan B, Islam Z et al. An interdependent network of functional enhancers regulates transcription and EZH2 loading at the INK4a/ARF locus. Cell reports. 2021. [PMID 33761351](https://pubmed.ncbi.nlm.nih.gov/33761351/). DOI: 10.1016/j.celrep.2021.108898. Access: abstract_metadata. Study: PMID33761351.

66. Saintamand A, Vincent-Fabert C, Marquet M et al. Eμ and 3'RR IgH enhancers show hierarchic unilateral dependence in mature B-cells. Scientific reports. 2017. [PMID 28348365](https://pubmed.ncbi.nlm.nih.gov/28348365/). DOI: 10.1038/s41598-017-00575-0. Access: abstract_metadata. Study: PMID28348365.

## Source-specific evidence cards

### [1] Master transcription factors and mediator establish super-enhancers at key cell identity genes.

**Shows:** High Mediator domains mark identity genes; factor depletion preferentially reduces their output.

**Does not show:** Calling is operational; global factor loss and reporter strength do not isolate class-specific synergy.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Results:paragraph4, 1:Results:paragraph5, 1:Results:paragraph11, 1:Results:paragraph12. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [2] Super-enhancers in the control of cell identity and disease.

**Shows:** High occupancy clusters associate with identity and disease loci across tissues.

**Does not show:** Nearest-gene assignment and mark enrichment do not establish a functional class or identity necessity.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:RESULTS:paragraph7, 1:RESULTS:paragraph8, 1:RESULTS:paragraph9. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [3] Convergence of developmental and oncogenic signaling pathways at transcriptional super-enhancers.

**Shows:** Prdm14 E3 deletion reduces neighboring acetylation; constituent effects and signaling responses vary.

**Does not show:** Single deletions and pathway stimulation do not distinguish emergent class behavior from local hierarchy.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:RESULTS:paragraph2, 1:RESULTS:paragraph3, 1:RESULTS:paragraph4. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [4] Genetic dissection of the α-globin super-enhancer in vivo.

**Shows:** Single and two selected double deletions show dominant activators, apparent additivity and retained domain structure.

**Does not show:** Later reconstruction reveals interactions missed by this limited deletion series; erythroid expression is not identity conversion.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 0:Results:paragraph18, 0:Results:paragraph20, 0:Results:paragraph21, 0:Results:paragraph22, 0:Results:paragraph25. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [5] Enhancers and super-enhancers have an equivalent regulatory role in embryonic stem cells through regulation of single or multiple genes.

**Shows:** Strong non-SE enhancers and dispensable called SEs coexist; cluster effects and redundancy vary.

**Does not show:** Selected loci and incomplete combinations do not establish every enhancer's equivalence or a universal additive law.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 0:Results:paragraph7, 0:Results:paragraph11, 0:Results:paragraph12, 0:Results:paragraph17. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [6] Hierarchy within the mammary STAT5-driven Wap super-enhancer.

**Shows:** Tested pairs are additive; combined proximal TF-site loss prevents cluster assembly and Wap induction.

**Does not show:** Motif and stage specificity matter; high milk-gene output is not a whole-lineage identity switch.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 0:RESULTS:paragraph7, 0:RESULTS:paragraph9, 0:RESULTS:paragraph10. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [7] Dissecting super-enhancer hierarchy based on chromatin interactions.

**Shows:** Hub repression and deletion disrupt output and neighbor marks more than nonhub deletion; CTCF-site effects are smaller.

**Does not show:** Transformed cells and selected clusters do not provide activity-matched conventional-cluster or whole-identity comparisons.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Results:paragraph21, 1:Results:paragraph23, 1:Results:paragraph24, 1:Results:paragraph25. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [8] Super-enhancers include classical enhancers and facilitators to fully activate gene expression.

**Shows:** Enhancers require interchangeable position-dependent facilitators for full output; reconstructed alleles rescue function.

**Does not show:** Erythroid populations retain lineage identity; same-locus platform cannot establish a class-wide distinct mechanism.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:RESULTS:paragraph4, 1:RESULTS:paragraph5, 1:RESULTS:paragraph13, 1:RESULTS:paragraph14, 1:RESULTS:paragraph15, 1:RESULTS:paragraph20, 1:RESULTS:paragraph21. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [9] The α-globin super-enhancer acts in an orientation-dependent manner.

**Shows:** Cluster inversion redirects output unlike single R2 inversion; boundary and promoter controls exclude simple insulation.

**Does not show:** Inversion includes genes and positional changes; no block in erythroid differentiation; linked correction body unexamined.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Results:paragraph4, 1:Results:paragraph6, 1:Results:paragraph8, 1:Results:paragraph12, 1:Results:paragraph16. No complete-paper or supplement audit claimed. ErratumIn: PMID40108171. Subclaims: SC2, SC3.

### [10] Transcriptional regulation and chromatin architecture maintenance are decoupled functions at the Sox2 locus.

**Shows:** Two key enhancer regions control output while distributed architecture persists; moderate engineered contact loss preserves RNA.

**Does not show:** Residual contacts remain; snapshot active-site fraction is not directly measured burst frequency.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 0:Results:paragraph4, 0:Results:paragraph7, 0:Results:paragraph11, 0:Results:paragraph17, 0:Results:paragraph18. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [11] A Myc enhancer cluster regulates normal and leukaemic haematopoietic stem cell hierarchies.

**Shows:** Cluster loss arrests differentiation; individual modules contribute combinatorially and additively.

**Does not show:** Original body and linked erratum unexamined; transformation and normal hierarchy are separate scopes.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. ErratumIn: PMID29769714. Subclaims: SC2, SC3.

### [12] Temporal dissection of an enhancer cluster reveals distinct temporal and functional contributions of individual elements.

**Shows:** Time-specific constituent roles and intronic amplifier yield super-additive Fgf5 induction.

**Does not show:** Expression induction does not uniquely demonstrate a distinct SE class or identity conversion caused by synergy.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [13] Transcriptional regulators of the Golli/myelin basic protein locus integrate additive and stealth activities.

**Shows:** Single and combined deletions show largely additive effects with lineage-hidden contributions.

**Does not show:** Shared protein output and lineage context do not isolate class-wide behavior; whole-body text unexamined.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [14] The interdependence of mammary-specific super-enhancers and their native promoters facilitates gene activation during pregnancy.

**Shows:** Native Wap promoter plus enhancer gives much larger induction than enhancer with heterologous promoter.

**Does not show:** Promoter compatibility explains local synergy; not proof that high-ChIP enhancer clusters are a distinct class.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [15] Genetic Dissection of a Super Enhancer Controlling the Nppa-Nppb Cluster in the Heart.

**Shows:** Cluster deletion reduces cardiac expression and causes enlarged hearts; promoters compete rather than cooperate.

**Does not show:** Physiological function is established at selected genes; does not show loss of cardiomyocyte identity or universal synergy.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [16] Defining Essential Enhancers for Pluripotent Stem Cells Using a Features-Oriented CRISPR-Cas9 Screen.

**Shows:** Selected native enhancers support self-renewal and pluripotency through defined target pathways.

**Does not show:** Screen hits include CREs without unique SE architecture; growth and identity endpoints require separation.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [17] Genome editing demonstrates that the -5 kb Nanog enhancer regulates Nanog expression by modulating RNAPII initiation and/or recruitment.

**Shows:** Native enhancer regulates polymerase recruitment or initiation with orientation/distance tests.

**Does not show:** One enhancer's contribution does not demonstrate emergent cluster synergy distinct from conventional enhancers.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [18] Super-enhancers conserved within placental mammals maintain stem cell pluripotency.

**Shows:** Conserved selected SE disruptions impair pluripotency; reporter activity tracks pluripotent state.

**Does not show:** Three selected conserved loci do not generalize all clusters; BRD4 is a broad coactivator.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [19] A flexible repertoire of transcription factor binding sites and a diversity threshold determines enhancer activity in embryonic stem cells.

**Shows:** TF-binding diversity supports strong enhancer activity with some master motifs dispensable.

**Does not show:** Sequence-complexity threshold is not a unique super-enhancer mechanism; synthetic behavior is scope-limited.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [20] Enhancer-promoter interactions can bypass CTCF-mediated boundaries and contribute to phenotypic robustness.

**Shows:** Strong clusters overcome barriers in epiblast/neural tissues; foregut loss yields specific phenotype.

**Does not show:** Boundary resistance and cell-state dependence do not establish a class-specific emergent identity code.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [21] Active and Inactive Enhancers Cooperate to Exert Localized and Long-Range Control of Gene Regulation.

**Shows:** Active and inactive enhancers cooperate in local and long-range lineage-specific regulation.

**Does not show:** Cooperation is broader than high-ChIP classification; no matched class-wide synergy comparison.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [22] Progressing super-enhancer landscape during mammary differentiation controls tissue-specific gene regulation.

**Shows:** Pregnancy seed enhancer becomes dispensable during lactation, revealing stage-specific compensation.

**Does not show:** Related Wap platform; distinct developmental output does not establish fixed cluster-class logic.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [23] An autoregulatory enhancer controls mammary-specific STAT5 functions.

**Shows:** Local enhancer loss lowers mammary Stat5 and lineage-dependent transcription.

**Does not show:** Single feedback enhancer and global downstream signaling do not isolate SE-specific identity necessity.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [24] Chromatin organizer SATB1 controls the cell identity of CD4+ CD8+ double-positive thymocytes by regulating the activity of super-enhancers.

**Shows:** SATB1 loss changes DP identity alongside reduced SE activity and contacts.

**Does not show:** Organizer loss alters many features; local cluster causal specificity is not isolated.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [25] Core transcription regulatory circuitry orchestrates corneal epithelial homeostasis.

**Shows:** RUNX1 and SMAD3 depletion switches corneal cells toward epidermal-like programs.

**Does not show:** Core TF dependence and chromatin changes do not isolate super-enhancer state as the unique mediator.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [26] Dynamic Control of Enhancer Repertoires Drives Lineage and Stage-Specific Transcription during Hematopoiesis.

**Shows:** Constituent hierarchy and opposing enhancer roles accompany lineage-specific transcription.

**Does not show:** SLC25A37 findings are reused in Huang; not independent replication of that same locus.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [27] MyoD- and FoxO3-mediated hotspot interaction orchestrates super-enhancer activity during myogenic differentiation.

**Shows:** MyoD-FoxO3 cooperation organizes Myogenin hotspots during differentiation.

**Does not show:** Shared factor function and selected cluster do not establish an emergent class beyond hierarchy.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [28] Super-enhancer switching drives a burst in gene expression at the mitosis-to-meiosis transition.

**Shows:** A-MYB and SCML2 regulate stage-specific SE landscape and meiotic gene induction.

**Does not show:** Global regulators and changing cell composition do not isolate individual cluster synergy.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [29] Epigenetic Mechanisms Mediating Cell State Transitions in Chondrocytes.

**Shows:** Enhancer occupancy and histone states change with differentiation and metabolic programs.

**Does not show:** Mapping and temporal covariation alone do not establish causal enhancer-class distinction.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [30] Bromodomain protein BRD4 directs and sustains CD8 T cell differentiation during infection.

**Shows:** BRD4 perturbations impair effector differentiation and maintenance alongside enhancer activity.

**Does not show:** Broad protein dependency is not selective SE disruption; immune state outcomes do not isolate cluster architecture.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [31] Deletion of Mediator 1 suppresses TGFβ signaling leading to changes in epidermal lineages and regeneration.

**Shows:** Med1 loss changes TGF-beta signaling, epidermal lineages and regeneration.

**Does not show:** Mediator has broad functions; pathway phenocopy does not establish SE-specific mechanism.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [32] Vitamin D receptor cross-talk with p63 signaling promotes epidermal cell fate.

**Shows:** Vdr loss impairs interfollicular regeneration while preserving sebaceous repopulation.

**Does not show:** SE colocalization and TF cross-talk do not isolate enhancer-class causal mediation.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [33] RANKL-responsive epigenetic mechanism reprograms macrophages into bone-resorbing osteoclasts.

**Shows:** Selected enhancer RNAs and BATF affect NFATc1 expression and osteoclast differentiation.

**Does not show:** RNA and BET effects do not separate cluster architecture from transcription chemistry or broad factor action.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [34] ARID1A safeguards the canalization of the cell fate decision during osteoclastogenesis.

**Shows:** ARID1A and BRD9 antagonism affects differentiation and associated coactivator assemblies.

**Does not show:** Broad remodeler loss cannot uniquely establish liquid-state necessity or a separate SE class.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [35] Rewiring of 3D chromatin topology orchestrates transcriptional reprogramming in muscle fiber-type specification and transformation.

**Shows:** Tissue-specific enhancer networks and STARD7 pathway activity associate with fiber-state changes.

**Does not show:** Abstract reverses conventional EDL/soleus fiber descriptors; anatomical identities need original-text verification before publication.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [36] Nucleome Dynamics during Retinal Development.

**Shows:** Three deletion sublines lack bipolar neurons with functional visual deficits and preserved gross retinal size.

**Does not show:** Local developmental necessity does not establish class-wide emergent synergy; local rescue not shown in examined passages.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:RESULTS:paragraph15, 1:RESULTS:paragraph16, 1:RESULTS:paragraph17, 1:RESULTS:paragraph18. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [37] Rapid Enhancer Remodeling and Transcription Factor Repurposing Enable High Magnitude Gene Induction upon Acute Activation of NK Cells.

**Shows:** Stimulus and lineage TFs redeploy at remodeled enhancers during acute induction.

**Does not show:** Primed versus de novo enhancer effects do not define a unique high-ChIP functional class.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [38] Neuroblastoma is composed of two super-enhancer-associated differentiation states.

**Shows:** PRRX1 changes expression and SE landscapes between mesenchymal and adrenergic states.

**Does not show:** Transcription-factor sufficiency and tumor heterogeneity do not isolate class-specific enhancer-state causality.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [39] Enhancer reprogramming in PRC2-deficient malignant peripheral nerve sheath tumors induces a targetable de-differentiated state.

**Shows:** Local enhancer gains and editing contribute to a neural-crest-like tumor state.

**Does not show:** Cancer genotype and broad drugs limit transfer to normal identity and emergent class distinction.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [40] Targeted brachyury degradation disrupts a highly specific autoregulatory program controlling chordoma cell identity.

**Shows:** Targeted TF degradation disrupts autoregulation more selectively than global transcriptional CDK inhibition.

**Does not show:** Protein specificity is not a selective liquid-state or enhancer-class intervention.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [41] A NOTCH1-driven MYC enhancer promotes T cell development, transformation and acute lymphoblastic leukemia.

**Shows:** One MYC regulatory element is required in thymocyte development and induced leukemia.

**Does not show:** Single enhancer within a broad region does not prove unique class mechanics; cancer and normal outcomes differ.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [42] The hegemonic EWSR1::ETS oncoprotein overrules core regulatory circuitry principles in Ewing sarcoma.

**Shows:** Predicted master regulators lack reciprocal circuit dependencies; the fusion protein dominates the state.

**Does not show:** Tumor-specific hierarchy is not a universal refutation of enhancer functions in normal identity.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [43] Therapeutic targeting of ependymoma as informed by oncogenic enhancer profiling.

**Shows:** Enhancer profiling identifies dependencies whose inhibition reduces tumor growth.

**Does not show:** Drug and target-gene dependence do not isolate enhancer-class synergy or normal-lineage identity determination.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [44] Identification of focally amplified lineage-specific super-enhancers in human epithelial cancers.

**Shows:** A MYC-associated constituent deletion or repression lowers MYC and oncogenic growth.

**Does not show:** Amplification changes dose; transformation dependence is a separate scope from normal identity and class distinction.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [45] CRISPR Screening of Transcribed Super-Enhancers Identifies Drivers of Triple-Negative Breast Cancer Progression.

**Shows:** Local PODXL enhancer and eRNA loss reduce output, migration and xenograft growth.

**Does not show:** Selected cancer enhancer dependence is not class-wide emergent synergy or normal cell-state conversion.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [46] Systematic discovery and functional dissection of enhancers needed for cancer cell fitness and proliferation.

**Shows:** Fitness enhancers have activating and repressive modules and context-specific accessibility.

**Does not show:** Fitness is not lineage identity; enhancer modularity extends beyond the SE label.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [47] A Genome-wide Framework for Mapping Gene Regulation via Cellular Genetic Screens.

**Shows:** Large-scale native screens link many distal elements to target output.

**Does not show:** Original method-plus-results report; broad repression may spread, and random combinations do not prove SE-wide synergy; erratum unread.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. ErratumIn: PMID30849375. Subclaims: SC2, SC3.

### [48] AP-1 subunits converge promiscuously at enhancers to potentiate transcription.

**Shows:** AP-1 hotspots predict lineage expression and signal response better than SE calls in tested analyses.

**Does not show:** Predictive comparison is not enhancer-class causality; shared factor recruitment remains a rival.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

### [49] Ligand dependent gene regulation by transient ERα clustered enhancers.

**Shows:** Ligand-dependent clustered enhancers control target output independently of called acetylation SEs.

**Does not show:** Transient signal response differs from persistent identity; local condensate association does not prove liquid-state causality.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [50] Hyperacetylated chromatin domains mark cell type-specific genes and suggest distinct modes of enhancer function.

**Shows:** Peak-breadth domains predict identity genes well; local dissections reveal roles obscured by SE calling.

**Does not show:** Alternative feature stratification does not identify one universal mechanistic class.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [51] Multiple structural maintenance of chromosome complexes at transcriptional regulatory elements.

**Shows:** Cohesin, condensin II and NIPBL enrich at SEs with associated transcription sensitivity.

**Does not show:** Complex depletion has architecture and cell-cycle effects; class-specific mechanism not isolated.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

### [52] Spatial enhancer clustering and regulation of enhancer-proximal genes by cohesin.

**Shows:** Loss weakens spatial enhancer clustering while chromatin marks and enhancer transcription persist.

**Does not show:** Broad cohesin effects and nearby expression changes do not uniquely prove high-ChIP cluster necessity.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [53] A tissue-specific self-interacting chromatin domain forms independently of enhancer-promoter interactions.

**Shows:** Domain formation persists without major enhancer interactions; internal contacts change.

**Does not show:** Same alpha-globin lineage platform; formation of a domain does not establish class-specific output mechanism.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [54] A Dynamic Folded Hairpin Conformation Is Associated with α-Globin Activation in Erythroid Cells.

**Shows:** Original computational predictions and structural data support a folded active-domain arrangement.

**Does not show:** Original model-plus-measurement research, not independent causal proof of thermodynamic phase separation.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [55] Interplay between CTCF boundaries and a super enhancer controls cohesin extrusion trajectories and gene expression.

**Shows:** Enhancer activation and extrusion trajectory are separable from paused boundary loops.

**Does not show:** Architecture and biochemical recruitment interact; selected locus does not define a separate enhancer class.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [56] Coactivator condensation at super-enhancers links phase separation and gene control.

**Shows:** BRD4 and MED1 assemblies and isolated domains support condensate plausibility.

**Does not show:** Chemical dissolution and in vitro droplets do not isolate native liquid-state necessity or unique SE class.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

### [57] Direct observation of a condensate effect on super-enhancer controlled gene bursting.

**Shows:** Nearby condensates predict enhanced bursting while basal bursts remain when distant.

**Does not show:** Spatial predictor is not isolated condensate-phase causality; linked correction body unexamined.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. ErratumIn: PMID38614098. Subclaims: SC2, SC3.

### [58] Single-cell 3D genome imaging shows super-enhancer hubs are rare, nonspecific, and neither necessary nor sufficient for transcriptional bursting.

**Shows:** Multiway SE hubs are rare, promiscuous and not a necessary burst configuration.

**Does not show:** Fixed snapshots do not test every pairwise interaction or disprove all local constituent cooperation.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [59] Mapping the evolving landscape of super-enhancers during cell differentiation.

**Shows:** Conserved, hierarchical and de novo clusters have distinct trajectories and constituent dependencies.

**Does not show:** Population heterogeneity and selected deletions do not prove shared emergent class identity mechanics.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

### [60] Selective inhibition of tumor oncogenes by disruption of super-enhancers.

**Shows:** BET inhibition preferentially reduces enhancer-associated oncogene output in tested tumor cells.

**Does not show:** Global coactivator and elongation effects can explain sensitivity without unique enhancer-class architecture.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [61] BET inhibition disrupts transcription but retains enhancer-promoter contact.

**Shows:** Output changes rapidly despite retained promoter contacts after inhibitor and degrader treatments.

**Does not show:** Broad chemistry and residual proteins remain; hexanediol is nonspecific and cannot uniquely test liquid state.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Results:paragraph7, 1:Results:paragraph12, 1:Results:paragraph17, 1:Results:paragraph21. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [62] Role of BRD4 in hematopoietic differentiation of embryonic stem cells.

**Shows:** BRD4 disruption impairs hematoendothelial potential without spontaneous ESC differentiation.

**Does not show:** Broad protein dependence and partly MYC-independent effects are not selective SE causality.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [63] Epigenetic Reader BRD4 (Bromodomain-Containing Protein 4) Governs Nucleus-Encoded Mitochondrial Transcriptome to Regulate Cardiac Function.

**Shows:** Genetic protein loss causes cardiac mitochondrial dysfunction unlike beneficial inhibitor results.

**Does not show:** Deletion and bromodomain inhibition test different functions; neither uniquely identifies SE architecture.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [64] Bromodomain-containing protein 4 (BRD4) regulates RNA polymerase II serine 2 phosphorylation in human CD4+ T cells.

**Shows:** BRD4-associated elongation and lineage output depend on bromodomain interactions.

**Does not show:** Enhancer and promoter biochemistry precedes the SE label and is not specific to that classification.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [65] An interdependent network of functional enhancers regulates transcription and EZH2 loading at the INK4a/ARF locus.

**Shows:** Conventional enhancer network is interdependent and can collapse after individual deletion.

**Does not show:** Direct comparator in principle, not an activity/number-matched experiment against selected SEs.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

### [66] Eμ and 3'RR IgH enhancers show hierarchic unilateral dependence in mature B-cells.

**Shows:** Paired regulatory regions show unilateral dependence in mature B cells.

**Does not show:** Locus-specific hierarchy and stage dependence do not distinguish emergent synergy from conventional enhancer cooperation.

**Access/status:** Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.