Research Papers
Hypothesis-aggregation papers generated by the AegisMind discovery engine. Each paper formalises hypotheses sourced from solver.press discoveries, presents a complete experimental validation package, and tracks confirmation status as experiments proceed.
mHTT Phase-Separated Condensates Sequester Transcription Factors in Huntington's Disease: A Flory-Huggins Computational Framework for Condensate-Disrupting Therapy
Mutant Huntingtin (mHTT) exon 1 undergoes polyQ-length-dependent liquid-liquid phase separation (LLPS) into gel-like condensates. We apply a Flory-Huggins polymer mixing framework as a two-point calibration of the published experimental phase boundary (Peskett et al. 2018), re-expressing those data in polymer-thermodynamic terms; the resulting polyQ-length-dependent phase diagram is illustrative rather than independently predictive. In this calibration, Q46-length protein crosses the phase boundary at a critical concentration of approximately 3.5 µM — physiologically accessible in HD striatal neurons. We propose transcription factor (TF) sequestration by condensate co-partitioning as a candidate downstream mechanism of mHTT toxicity. A sensitivity analysis shows the predicted magnitude is governed by the nuclear volume fraction of the mHTT phase, which current data do not constrain, rather than by the partition coefficients; at plausible volume fractions it falls one to two orders of magnitude below the mean −44.7% reduction in SP1/CBP/TFIID target gene expression reported in HD striatum. We hypothesize that BET bromodomain inhibitors (JQ1, OTX015) and mitoxantrone analogues can restore TF availability by disrupting mHTT condensates. Patent AU2026905785.
Convergence of Performative Scenario Optimization to Classical Stochastic Programming in the Vanishing-Feedback Limit
Performative prediction — the phenomenon whereby a deployed decision model influences the data distribution it is trained on — fundamentally distinguishes real-world optimization from classical stochastic programming (SP). We present two complementary computational hypotheses formalizing convergence of performatively stable solutions to classical SP optima as decision-feedback strength ε → 0. H₁: convergence rate is O(ε·L) where L is the Lipschitz modulus of the distribution map. H₂: the convergence exhibits an entropic regularization analogy analogous to entropic optimal transport converging to classical OT. H₁ is tested computationally across five synthetic problem families (linear-quadratic, portfolio allocation, newsvendor, logistic regression, quadratic programming): log-log slope α ∈ [1.000, 1.028] with R² ≥ 0.9995 in all cases confirms exact O(ε) convergence. Corrected September 2026 — the paper previously reported both hypotheses as validated. The pre-registered criterion C ≤ 1.5·L̂ for ≥3 of 5 families was met by 2 of 5 and failed. H₂ is not independent of H₁ (§2.4 relates them by ε ↔ 1−ε), and its one distinguishing prediction — α < 1 for non-smooth displacement — is refuted: discontinuous and non-differentiable maps both converge at α ≈ 1.000.
Equilibrium Computation, Matrix Interpolation, and Ergodicity-Onset Optimization for Ergotropy Protection in Open Quantum Batteries
SHELVED 4 August 2026; H₂ additionally withdrawn 21 August 2026. The programme proposed five computational hypotheses for protecting ergotropy in open quantum batteries. Two were reported as validated and neither holds, for the same reason. Both simulations initialised the qubit already excited with the cavity in vacuum — the script says so in a comment: "Initial state: qubit excited, cavity vacuum" — so no energy ever had to be transferred, and what was measured was how well a pre-loaded excitation is retained rather than how well a battery charges. H₁ claimed 84.9% improvement from Nash-equilibrium cavity detuning at Δ = −10g; re-run with a real charging phase (cavity charger, qubit starting in the ground state) the optimum inverts to exact resonance and Δ = −10g yields exactly zero ergotropy. H₂ claimed 54.7% improvement at an interpolated optimum of g* = 0.01; with κ = 0.10 that is g/κ = 0.1, and charging requires g/κ ≥ 3, so the identified optimum lies in the regime where the battery cannot charge at all. Both optima also sat at the edge of their sampled grids. The original parameters (g/κ = 1.0) could not charge at any detuning in any case, since transfer time π/2g = 15.7 exceeds cavity lifetime 1/κ = 10. The simulation and the statistics were sound; the initial state was not, and a p-value of 10⁻³⁵ on a correct simulation of the wrong setup is still worthless. H₃–H₅ were never tested and are withdrawn with the programme. This work was never deposited and has no DOI.
Quorum Sensing Loss-of-Function Mutations Impose Polymicrobial Fitness Costs: A Computational Hypothesis for Combined QS-Inhibitor and QS-Dependent Antibiotic Therapy
Antibiotic resistance mediated through quorum sensing (QS) loss-of-function creates a strategic vulnerability: QS-deficient mutants escape QS-dependent antibiotic action but simultaneously lose access to cooperative extracellular public goods, making them exploitable as cheaters in polymicrobial environments. We formalize the hypothesis that combined QS-inhibitor therapy with QS-dependent antibiotics creates a doubly unfavorable evolutionary landscape for resistant mutants — simultaneously imposing antibiotic selection pressure and ecological disadvantage. Using Lotka-Volterra public-goods competition models, we derive conditions under which selection coefficients against QS-deficient mutants are strongly negative under combined therapy but near-zero under antibiotic monotherapy. The hypothesis predicts selection coefficients s ≤ −0.05 per passage in ≥5 of 6 replicates under combined therapy, versus positive selection under monotherapy. We present a complete experimental validation package comprising a 160-day, three-phase study using isogenic QS-deficient mutants (ΔlasR, ΔrhlR, Δagr) in polymicrobial competition assays.
Evolutionary Traps in WHO Priority Pathogen Collateral Sensitivity Networks: A Graph-Theoretic Hypothesis and Experimental Validation Design
REFUTED 20 August 2026. The hypothesis was that the directed collateral sensitivity graph for WHO priority pathogens contains strongly connected components of size ≥3 defining closed evolutionary traps — resistance cycles with no viable single-step escape — which sequential cycling therapy could exploit to hold a pathogen susceptible indefinitely. It was tested against 104,337 susceptibility records from BV-BRC and a candidate was found: a 3-node component in K. pneumoniae linking imipenem, meropenem and tetracycline. It does not survive stratification by clonal lineage. Within MLST sequence types the association is absent (pooled odds ratios 0.93 and 0.95, Cochran-Mantel-Haenszel p = 0.73 and 0.80, confidence intervals excluding the unadjusted estimates), and the two dominant lineages pull in opposite directions (ST307, 1.26; ST258, 0.67). Carbapenem-resistant and tetracycline-resistant phenotypes co-occur because they are carried by different successful clones, not because resistance to one induces susceptibility to the other. The manuscript reporting the positive finding was withdrawn from journal review, and the experimental validation package below is withdrawn with it: an isogenic experiment is a within-lineage comparison by construction, so the prediction it would have tested has already been evaluated across 473 co-tested isolates and is not supported. The earlier claim of 50,000–200,000 deaths averted per year was conditional on confirmation and does not stand. What survives is methodological, and is reported separately: none of the checks that made this signal convincing — FDR correction, permutation significance, drug-class specificity, temporal stability — distinguishes a trade-off from linkage within a clone, and the controls that do are inexpensive.
Aggregated EVP packages
Papers are grouped into EVP clusters that share experimental infrastructure and compound the evidence for related mechanisms.