CEREBRO-X

Computational Drug-DDS Engineering · CEREBRO-X | Donepezil

H01 · BBB Crossing Animation — Donepezil + plga
5-stage simulation based on computed DLVO + transcytosis + endosomal escape data
IV Injection → Bloodstream
100.0%
BBB Crossing (with DDS)
36.8%
Native BBB Crossing
2.7×
Enhancement Factor
64%
Endosomal Escape Eff.
1.00
Stealth Index
130 nm
Carrier Size
H02 · PBPK-CNS Digital Twin — 6-Compartment Time Course
Donepezil + plga | Disease state: healthy | Radau ODE solver
0.2213
Cmax Brain (µg/mL)
2.4h
t_max Brain
1.06181
Kp,brain
7.297
AUC Brain (µg·h/mL)
6.87
AUC Plasma (µg·h/mL)
healthy
Disease State
H08 · Molecular Docking — RVG29 ↔ BBB Receptor Binding
Donepezil | ΔG_avidity = -10.1 kcal/mol | Kd = 78 nM (Moderate (10-100nM)) | LIE + Bell avidity model
-10.1
ΔG (kcal/mol)
78 nM
Kd (affinity)
42474
Ligands/particle
12882219.1s
Residence time
100.0%
BBB Enhancement
130nm
Carrier size
ParameterValueMethod
ΔG single ligand-12.50 kcal/molLIE approximation
ΔG avidity-10.09 kcal/molBell cooperative binding
Receptors engaged50Geometric model
Kd classModerate (10-100nM)Equilibrium binding

Ligand-receptor docking uses Linear Interaction Energy (LIE) approximation. Avidity computed via Bell 1978 multi-valent model. Higher avidity = longer residence = more transcytosis events.

H09 · Docking + Release — Combined Delivery Simulation
Donepezil from plga | ΔG=-10.1 kcal/mol | t50_blood=23.2h | Escape=64%

Docking affinity → release kinetics correlation

Endosomal escape efficiency vs release rate

H03 · Drug Release Profile — Donepezil
First Order model | pH-triggered (endosomal) | Max EE = 75%
23.2h
t50 in Blood
36.2h
t50 Endosomal
75%
Max Release (EE)
H14 · Shelf-Life & Degradation Predictor — Arrhenius Kinetics
Donepezil | Grade: MARGINAL (3-6 months) | Dominant: Drug leakage | ICH Q1A standard
122d
t90 Shelf-Life
MARGINAL
Quality Grade
Drug leakage
Main Degradation
H17 · Glymphatic Clearance — Sleep/Wake Cycle Animation
Donepezil | t½ waking=12.4h | ECM binding=0.74 | GOOD: particle size > ECM pore -- extended CNS retention

Purple bands = sleep (00:00–08:00), 3.5× faster clearance (Xie 2013, Science). ECM binding index 0.74: particles trapped in ECM — extended retention.

H05 · DDS Ranking Dashboard — Top 20 Formulations
Ranked by Composite Score (Drug+DDS biophysics combined)
#1 RVG29-PLGA-NP
H10 · Regression Docking — Composite Score vs BBB Enhancement
Donepezil | All 8 DDS formulations | Hover for details
H11 · Efficiency Heatmap — DDS × Performance Matrix
Donepezil | Top 8 DDS | Normalized 0-1 (green=high, red=low)
H13 · DDS Comparison Radar — Top-5 Formulations
Multi-dimensional comparison across key biophysical metrics (all metrics scaled to 0-100)
H06 · DLVO Colloidal Stability — Potential Energy Curve
plga | Zeta=-22.0 mV | Size=130 nm | V_total=5.8 kT | ⚠️ UNSTABLE
5.8 kT
V_total (>25kT = stable)
-22.0 mV
Zeta Potential
UNSTABLE
Colloidal Status

DLVO theory: V_total = V_vdW (attraction) + V_EDL (electrostatic repulsion). Threshold 25kT prevents aggregation in blood. Debye length = 0.78 nm (physiological ionic strength).

H07 · Score Contribution Breakdown — DDS Score Waterfall
Donepezil | RVG29-PLGA-NP | Composite Score = 80.5

Illustrative decomposition of the composite score into named per-feature terms (a fixed heuristic formula, not Shapley values from a trained model). Green = positive contribution. Red = negative.

H12 · Off-Target QSAR Heatmap — 50-Receptor Panel
Donepezil | 20 high-risk targets | CRITICAL
20
High-Risk Targets
⚠️
Cardiac Safety
⚠️
Hepatic Safety
H18 · Microglial Activation — Neuroinflammation Risk
Donepezil | Score=0.043 | LOW -- safe for CNS application
4%
Neuro-inflammation Score
1.4×
IL-6 fold-change
1.3×
TNF-α fold-change
0.100
TLR2/4 score

→ No mitigation needed

H15 · Synthetic Clinical Trial — N=500 Virtual Patients
Donepezil | Monte Carlo simulation | Pharmacogenomics-adjusted
NO-GO / REFORMULATE
Go/No-Go Decision
100.0%
Overall Response Rate
0.07 mg/kg
Optimal Dose
100.0%
Young Adults (<65)
100.0%
Elderly (>65)
91.2%
Severe AE (<5% = GO)
H16 · CEREBRO-X Capability Radar — 50 Active Science Modules
Donepezil | 8 capability dimensions across all 62 science module outputs
Capability DimensionScoreModules Active
BBB Engineering100%✅
PK Prediction95%✅
Safety Screening90%✅
Manufacturing85%✅
Regulatory (FDA)90%✅
Clinical Translation80%✅
IP & Economics85%✅
Stability90%✅
50
Active modules / 62 total science points
H20 · Bootstrap Validation — 95% Confidence Intervals
1000 resamples of each formulation's 8 real principle-group scores | Uncertainty quantification for score ranking
H21 · Drug Problems → DDS Solutions Mapping — Donepezil
Automatically identified delivery barriers and how the DDS resolves each
SeverityProblem IdentifiedEvidenceDDS SolutionExpected Outcome
HIGH Cardiac off-target risk (hERG/Nav/Cav channels) QSAR panel: cardiac receptor score > safety threshold CNS-targeted plga reduces systemic expos CNS bioavailability 53.0% with reduced c
H19 · Lyophilization Cycle Optimizer
Donepezil | Tg'=-30.0°C | Primary drying=-32.0°C | Total=42h | OK: primary drying fixed at Tg'-2degC safety margin (-32 degC)
StepSetting
Freezing-50°C @ 1°C/min
Primary drying-32°C, 0.148 mbar
Secondary drying+25°C, 0.05 mbar
Tg' safety margin-2.0°C above Tg'
Cake collapse riskOK: primary drying fixed at Tg
42h
Total cycle time
-30°C
Tg' (glass transition)
H24 · Cryo-Chain Thermal Excursion — Donepezil
Lipid phase transition thermodynamics | Arrhenius leakage model
RELEASE BATCH
Batch Decision
75.0% → 74.2%
EE Before → After
0.0%
Lipid Melted

Excursion Temperature: -20°C for 4h
Lipid Tm: 42°C
ΔT above Tm: -62.0°C
Leakage rate: 0.000100 /h
Analytical check: Re-test DLS particle size + HPLC drug content before release

H26 · FUS-Responsive Nanocarrier — Acoustic BBB Opening
Donepezil | 0.5 MHz | MI=0.40 | BBB open 42 min | Inject 130 nm DDS 5 min before FUS. FUS opens BBB for 42 min. Predicted 48% enha
0.5 MHz
FUS Frequency
0.40
Mechanical Index
283 kPa
Negative Pressure
42 min
BBB Open Window
48%
FUS-Enhanced Uptake
100%
Standard BBB Enh.

FUS + microbubbles open BBB via stable cavitation (MI 0.3-0.5). Inertial cavitation (MI>0.8) = irreversible damage. Carrier injected 5 min before FUS for optimal uptake.

Structural integrity: OK

H23 · In-Silico Biodistribution — Organ Map
Donepezil | CNS/off-target ratio = 1.128 | No animal studies needed (3R principle)

🧬 All 62 Principles — Unified Scoreboard (Top-1 DDS)

For Donepezil, the Top-1 DDS achieved a composite score of 80.5/100 (EXCELLENT). Below are the CNS group rollups followed by the FULL 62-principle scoreboard covering all three pipeline classes:  A (Surrogate)  B (Deep)  C (Translational)

CNS Principle Group Rollups

G1 CNS Delivery: 87.1/100
G2 Release Kinetics: 54.4/100
G3 Stability: 76.5/100
G4 Safety: 85.5/100
G5 Glymphatic BBB: 61.7/100
G6 Manufacturability: 77.8/100
G7 DrugDDS Fit: 61.5/100
G8 Translational: 0.0/100

All 62 Principles (P01 → P62)

P# Class Title Score Conf. Method Reference
P01BAdversarial Stress-Testing Engine62.4MODERATEMean of 6 stress scenarios (pH4.5, pH2, pH8.5, 42°C, complement, oxidation); each sce
P02BCross-Species PK Scaling (Allometric)100.0MODERATEBW^0.75 scaling + class-specific adjustment
P03ACompetitive DDS Landscape Radar81.8LOWNovelty = 100 - frequency_of_combo_in_CNS_trials, modified by whether the drug's LogP
P04BQuantum Coherence Transport Model0.7MODERATEWKB tunneling: P ∝ exp(-2·barrier)
P05AIn-Silico Patient Subgroup Stratifier50.0MODERATE% subgroups responding (lower CYP risk = more universal)
P06ALysosomal Trafficking Predictor77.5MODERATECarrier endosomal_escape ⊕ drug-passive permeation at pH 5.5; drug-passive = LogP-dri
P07AReal-Time Literature Mining (PubMed)100.0LOWlog-count proxy of literature support
P08BDegradation Kinetics Under Oxidative Stress57.3MODERATEArrhenius k=A·exp(-Ea/RT) for carrier; minus penalties for drug-side oxidation-prone
P09ADigital Pharmacovigilance Engine100.0MODERATE100 - sum(metabolite organ accumulation risks)
P10BLNP Ionization State Predictor60.0LOWNon-LNP carrier — falls back to drug-side ionization stability at storage pH 7.4 (low
P11BFormulation Instability Fingerprint90.0MODERATE100 - sum(weak-bond penalties from SMILES)
P12BCNS Disease-Stage-Aware Dosing78.8MODERATEScore = 100·BBB_integrity_factor × (0.4 + 0.6·CNS-MPO/6) + targeting_boost (larger bo
P13BPBPK Digital Twin100.0MODERATEAUC_brain/AUC_plasma proxy; 3-compartment surrogate
P14AIn-silico Dissolution & Release Profile82.6MODERATEt50 from carrier kinetics × drug-membrane partition factor (10^logK_mem); HBD-anchori
P15AShelf-life & Degradation Predictor92.5MODERATEArrhenius baseline × (0.7 + 0.3·EE) × drug_hydrolysis_factor (0.08 penalty per hydrol
P16BScale-up & Manufacturability45.7MODERATEScale-readiness × shear-robustness, minus drug-processability penalties (MW, HBD-crys
P17ANanotoxicity & Immunogenicity Screening86.2MODERATEMean of (hemolysis penalized by drug net+ charge, complement penalized for biologics,
P18BActive Targeting & Receptor Binding87.2HIGHLigand-affinity table × density × drug-compatibility (receptor-pathway-specific drug
P19AQbD - Quality by Design Engine100.0MODERATEFraction of CQAs (size, zeta, EE, PDI, drug-loading) within ICH spec — 5th CQA depend
P20ACost-Efficiency Engine57.4LOWScore = 100 - 3·($/mg estimate, drug+carrier+ligand+API). Drug API cost from stereoce
P21CPre-IND Regulatory Reports0.0—skipped_deep_validation_insufficient
P22AProtein Corona Predictor61.5MODERATE100 × exp(-corona_thickness/10); thickness from carrier (|zeta|, size) + drug-side (L
P23BDynamic Crystal Polymorphism77.0MODERATE100 - polymorph risk (rot_bonds + H-bond pattern)
P24BShear-Stress & Scale-Up Collapse89.3MODERATElog10(crit_shear/operating_shear) × 30; crit_shear softened by drug-load for lipid ca
P25AExtractables & Leachables80.0MODERATE100 - leachables risk by drug LogP × carrier
P26AMicrobiome-Excipient Interactions81.1MODERATE100 - mean_microbiome_degradability(carrier) - drug-side risk (anionic-fraction, pola
P27ALyophilization Cycle Optimizer77.5MODERATE100 - (|Tg'| - 5) × 1.5
P28A3D-Printed Polypill Rheology80.0MODERATECarrier-specific printability index
P29BBiomimetic & Exosome Engineering55.0MODERATEStealth-from-macrophage score; PEG ≥5% boost
P30BQM/MM Stimuli-Responsive Cleavage45.0MODERATEScore = 50 × |7.4 - pH_trigger|
P31BIn-Silico Biodistribution100.0MODERATEbrain% = BBB · ligand · size_match · 100; 100 if ≥5%
P32CAutomated FTO & IP Evader0.0—skipped_deep_validation_insufficient
P33BBBB Quantum Breaker (Trojan-Horse Design)56.4MODERATE(0.6·ligand + 0.4·size) × density_factor
P34ADNA Logic Gates & Bio-computing20.0MODERATEDNA-based carriers score high
P35AMicrogravity Formulation Engine100.0LOWSedimentation-Péclet proxy; smaller = better
P36AGeopolitical Supply-Chain Resilience60.0MODERATESupplier-diversity index by carrier + ligand
P37AEco-Destructible Pharma95.0MODERATEBiodegradability index by carrier class
P38BGlymphatic Clearance Trap100.0MODERATEStokes-Einstein: optimum 80-150 nm
P39AMicroglial Activation & Neuroinflammation100.0MODERATE100 - (cationic_charge + carrier_risk - PEG_protection)
P40BIntranasal-to-Brain Delivery65.0MODERATEMucoadhesion + thermo-responsive boost
P41BExosome Cargo Loading Thermodynamics40.0LOWNot an exosome carrier — N/A
P42ARegion-Specific Spatiotemporal Navigation85.0MODERATELigand-region specificity table
P43BFUS-Responsive Nanocarriers20.0MODERATEFUS-response by carrier acoustic properties
P44BCNS-Specific PBPK Time-Machine100.0MODERATEAUC over 24h therapeutic window proxy
P45CFDA 21 CFR Part 11 Compliance0.0—skipped_deep_validation_insufficient
P46APolypharmacy & DDI Simulator100.0MODERATECYP-inhibition heuristic from SMILES
P47BFree Energy Perturbation (FEP+)90.0LOWVina-like ΔG proxy from MW + LogP (deep mode runs full FEP+)
P48AOff-Target Toxicity & QSAR (50-receptor)55.0MODERATE50-receptor QSAR surrogate (hERG, 5HT2B, AhR risks)
P49AOrgan-on-a-Chip Simulator80.0MODERATEMicrofluidic compatibility (size + PDI)
P50BCryo-Chain Thermal Excursion Predictor71.5MODERATEDistance from -20°C phase transition × 1.3
P51BTerminal Sterilization Survivability85.0MODERATECarrier gamma-radiation survival (25 kGy)
P52AContinuous Manufacturing Digital Twin80.0MODERATEContinuous-process readiness by carrier
P53ADark Data & Negative Results Vault100.0MODERATESimilarity-to-documented-failure
P54APharmacogenomic-Guided Targeting70.0MODERATEClass-based pharmacogenomic applicability
P55CAutomated Grant & NIH Proposal Generator0.0—skipped_deep_validation_insufficient
P56CPatentability Score Engine0.0—skipped_deep_validation_insufficient
P57BMicrofluidics & LNP Synthesis Digital Twin90.0MODERATEMicrofluidic readiness for size 50-150 nm
P58BImpurity Cascade Predictor85.0MODERATECarrier residual-metals impurity profile
P59B4D Shape-Shifting Carriers30.0MODERATEStimuli-responsive bonus by release kinetics
P60ASwarm Nanorobotics Intelligence15.0LOWSwarm capability (most carriers: no)
P61BSynthetic Clinical Trials & Virtual Humans95.0MODERATEVirtual-cohort responder fraction
P62ABiobetter / Supergeneric Generator50.0LOWNovelty distance from on-market CNS DDS

RVG29-PLGA-NP: composite CNS-principle score = 80.5/100 (EXCELLENT). Carrier: plga, size 0.0 nm, ζ +0.0 mV, ligand: RVG29. Strongest group: G1 CNS Delivery (87.1/100); weakest: G8 Translational (0.0/100). Evaluated against all 57 Class A surrogate principles.

🔬 H28 — Class B Deep Physics Validation (Top-1 DDS)

Verdict: MARGINAL   |   19/28 principles validated (67.9% — threshold 70%)

Combined score (surrogate pass-through + real physics): 19/28 principles scored PASSED (67.9%). Of these, only 7/28 principles ran independent deep computation (passed 7/7 = 100.0%); the rest re-used their Class-A surrogate score pending a future full-physics HPC run. Cite independent_pct, not pct, as evidence of physics-based validation.

PrincipleValidatedScoreValue ConfidenceMethodNarrative
P01✅62.462.41MODERATEFull MD stress simulation — surrogate value confirmed (full-physics HPC deferred)Full MD stress simulation: surrogate score 62.41 PASSED. Full deep simulation requires external HPC.
P02✅70.054.28MODERATEResolved clinical half-life (bundle tier 6); Mahmood 2007 exponents shown for reference only (half-lClinical t½ = 54.3h (population regression / class estimate, bundle tier 6). Reference allometric exponents (Mahmood 2007) for a genuine ani
P04❌0.70.0037LOWFull QM tunneling (QCElemental + ASE) — surrogate value confirmed (full-physics HPC deferred)Full QM tunneling (QCElemental + ASE): surrogate score 0.74 NEEDS REVIEW. Full deep simulation requires external HPC.
P08❌57.357.3LOWFull radical-chain reaction MD — surrogate value confirmed (full-physics HPC deferred)Full radical-chain reaction MD: surrogate score 57.3 NEEDS REVIEW. Full deep simulation requires external HPC.
P10✅60.060.0MODERATEConstant-pH MD simulation — surrogate value confirmed (full-physics HPC deferred)Constant-pH MD simulation: surrogate score 60.0 PASSED. Full deep simulation requires external HPC.
P11✅90.090MODERATEDFT bond dissociation energies — surrogate value confirmed (full-physics HPC deferred)DFT bond dissociation energies: surrogate score 90 PASSED. Full deep simulation requires external HPC.
P12✅78.878.75MODERATEStage-specific PBPK with full BBB physiology — surrogate value confirmed (full-physics HPC deferred)Stage-specific PBPK with full BBB physiology: surrogate score 78.75 PASSED. Full deep simulation requires external HPC.
P13✅100.012.6473HIGH3-compartment PBPK ODE (scipy.integrate.odeint): blood ⇌ brain ⇌ peripheral, 24h time course. k_bb_iBrain AUC = 1813.87 units·h, plasma AUC = 143.42 units·h, ratio = 12.6473. Cmax_brain = 78.74 at t = 8.7h. PASSED AUC ratio.
P16❌45.745.72LOWCFD of 1000L bioreactor — surrogate value confirmed (full-physics HPC deferred)CFD of 1000L bioreactor: surrogate score 45.72 NEEDS REVIEW. Full deep simulation requires external HPC.
P18✅98.0-9.8HIGHMM/GBSA-style ΔG estimate from validated ligand-receptor Kd database. Full atomistic docking deferreLigand 'rvg29' → receptor 'nAChR', ΔG = -9.8 kcal/mol. STRONG BBB-targeting.
P23✅77.077.0MODERATECSP via DFT — surrogate value confirmed (full-physics HPC deferred)CSP via DFT: surrogate score 77.0 PASSED. Full deep simulation requires external HPC.
P24✅89.389.31MODERATEFull CFD + MD coupling — surrogate value confirmed (full-physics HPC deferred)Full CFD + MD coupling: surrogate score 89.31 PASSED. Full deep simulation requires external HPC.
P29❌55.055LOWFull membrane-fusion MD — surrogate value confirmed (full-physics HPC deferred)Full membrane-fusion MD: surrogate score 55 NEEDS REVIEW. Full deep simulation requires external HPC.
P30❌45.045.0LOWFull QM/MM (PySCF or ORCA) — surrogate value confirmed (full-physics HPC deferred)Full QM/MM (PySCF or ORCA): surrogate score 45.0 NEEDS REVIEW. Full deep simulation requires external HPC.
P31✅100.054.05MODERATE7-organ whole-body distribution with size + charge + ligand-status modifiers.Brain uptake = 54.05%, liver = 7.5%, spleen = 3.6%. GOOD CNS DELIVERY.
P33❌56.456.4LOWAtomistic docking + SMD pulling — surrogate value confirmed (full-physics HPC deferred)Atomistic docking + SMD pulling: surrogate score 56.4 NEEDS REVIEW. Full deep simulation requires external HPC.
P38✅95.415.6HIGHStokes-Einstein D = k_B·T/(6πηr) at 37°C, η_CSF = 7×10⁻⁴ Pa·s. Brain residence scaled to 50-nm referD_CSF = 4.99 µm²/s, brain residence ≈ 15.6h. IDEAL (target 6-48h).
P40✅65.065MODERATEFull nasal cavity CFD — surrogate value confirmed (full-physics HPC deferred)Full nasal cavity CFD: surrogate score 65 PASSED. Full deep simulation requires external HPC.
P41❌40.040LOWFull membrane mechanics MD — surrogate value confirmed (full-physics HPC deferred)Full membrane mechanics MD: surrogate score 40 NEEDS REVIEW. Full deep simulation requires external HPC.
P43❌20.020LOWFull acoustic radiation force MD — surrogate value confirmed (full-physics HPC deferred)Full acoustic radiation force MD: surrogate score 20 NEEDS REVIEW. Full deep simulation requires external HPC.
P44✅100.047.76HIGH4-compartment CNS-PBPK ODE with glymphatic clearance: Blood ⇌ Brain_ECF → CSF → Blood. Stokes-EinsteTherapeutic window = 47.8h (brain ≥ 5% dose). Brain AUC = 1278.0, Cmax = 72.91. PASSED.
P47✅84.6-7.69HIGH — real AutoDock Vina docking on experimental receptor structureAutoDock Vina 1.2.7 (real docking)Predicted ΔG_binding = -7.69 kcal/mol (AutoDock Vina 1.2.7 (real docking)). MODERATE affinity.
P50✅71.571.5MODERATEFull coarse-grained MD lipid phase transition — surrogate value confirmed (full-physics HPC deferredFull coarse-grained MD lipid phase transition: surrogate score 71.5 PASSED. Full deep simulation requires external HPC.
P51✅85.085MODERATEFull radical-chain damage MD — surrogate value confirmed (full-physics HPC deferred)Full radical-chain damage MD: surrogate score 85 PASSED. Full deep simulation requires external HPC.
P57✅90.090MODERATEFull CFD of mixer geometry — surrogate value confirmed (full-physics HPC deferred)Full CFD of mixer geometry: surrogate score 90 PASSED. Full deep simulation requires external HPC.
P58✅85.085MODERATEFull QM/MM cascade simulation — surrogate value confirmed (full-physics HPC deferred)Full QM/MM cascade simulation: surrogate score 85 PASSED. Full deep simulation requires external HPC.
P59❌30.030LOWFull coarse-grained MD morphological transition — surrogate value confirmed (full-physics HPC deferrFull coarse-grained MD morphological transition: surrogate score 30 NEEDS REVIEW. Full deep simulation requires external HPC.
P61✅95.095MODERATEFull Monte-Carlo population PBPK — surrogate value confirmed (full-physics HPC deferred)Full Monte-Carlo population PBPK: surrogate score 95 PASSED. Full deep simulation requires external HPC.

📋 H29 — Class C Translational Deliverables (Top-1 DDS)

Translational outputs for Donepezil's validated Top-1 DDS. These are NOT used for ranking — they are administrative deliverables triggered only AFTER deep physics validation passes.

P21: Pre-IND Outline

Status: skipped_deep_validation_insufficient
Score: —

Deep validation passed only 19/28 principles. Translational deliverables withheld until a higher-ranked DDS passes deep validation.

P32: Freedom-to-Operate

Status: skipped_deep_validation_insufficient
Score: —

Deep validation passed only 19/28 principles. Translational deliverables withheld until a higher-ranked DDS passes deep validation.

P45: 21 CFR Part 11

Status: skipped_deep_validation_insufficient
Score: —

Deep validation passed only 19/28 principles. Translational deliverables withheld until a higher-ranked DDS passes deep validation.

P55: Grant Outline

Status: skipped_deep_validation_insufficient
Score: —

Deep validation passed only 19/28 principles. Translational deliverables withheld until a higher-ranked DDS passes deep validation.

P56: Patentability

Status: skipped_deep_validation_insufficient
Score: —

Deep validation passed only 19/28 principles. Translational deliverables withheld until a higher-ranked DDS passes deep validation.

🔁 H30 — Top-N Fallback Audit Trail

For Donepezil: each candidate DDS that was tried in the Class B deep validation phase. If the Top-1 fails the 70% threshold, we fall back to Top-2, then Top-3 — with explicit reasons recorded.

RankDDSSurrogateVerdict Promoted?Failure ReasonTransition Reason
#1 RVG29-PLGA-NP 80.49 MARGINAL
67.9% (19/28)
— Deep validation MARGINAL: only 19/28 principles validated (67.9%, threshold 70%). Failed principles include: P04, P08, P16, P29, P30. Falling back to rank #2 because this DDS did not meet the 70% deep-validation threshold.
#2 Tf-SLN-Stealth 79.24 MARGINAL
60.7% (17/28)
— Deep validation MARGINAL: only 17/28 principles validated (60.7%, threshold 70%). Failed principles include: P01, P04, P08, P16, P24. Falling back to rank #3 because this DDS did not meet the 70% deep-validation threshold.
#3 ApoE-SLN 79.15 MARGINAL
60.7% (17/28)
— Deep validation MARGINAL: only 17/28 principles validated (60.7%, threshold 70%). Failed principles include: P01, P04, P08, P16, P24. No more candidates in Top-3 — reverting to rank #1 and reporting with MARGINAL verdict.