Enter 2-3 SMILES with concentration sliders • Sweetener potency-corrected • E-tongue reference comparison • Full XAI reasoning
💊 Taste Masking Assessment (Pharma)
ASTREE-equivalent placebo matching • Euclidean distance in 7-sensor space • Masking efficiency score
🔎 Adulteration Detection
Detect contamination via sensor profile shift • Like ASTREE orange juice analysis • Detection threshold calculation
0.0010.11.010100
🏆 Attribute Ranking
Rank compounds on individual taste dimensions • Like ASTREE sourness/sweetness ranking • Competitive benchmarking
⚙ E-Tongue Calibration & Anomaly Detection
Standard Taste Kit calibration • Salty/sour null zone • Umami/salt synergy • Water drift • Temperature correction • Negative sweetness
📈 Concentration Response Curve
Standard addition method • Sensor response vs concentration across all 7 channels • ASTREE equivalent
🌐 PCA Taste Map — ASTREE-Format Discrimination
Principal Component Analysis of 7-sensor responses • Discrimination Index (DI ≥ 94% = validated) • Per-attribute intensity ranking • SRCS classification • GC-MS alignment
ASTREE Application Presets
Drug Cross-Reactivity & False Positive Detector
SnO₂ MOX fingerprint superposition • CODEX frequency combination • 11 drug panel screening • GRE Pillar integration
When two or more compounds are present, their combined frequency signatures can mimic a target drug — causing FALSE POSITIVES on immunoassay drug tests.
CASP16 Pharmaceutical Ligand Benchmark
CODEX Genesis Resonance vs World's Best — No ML, No Training Data, No Templates
Affinity ranking on CASP16 pharmaceutical drug discovery targets (Roche + Idorsia, 2024)
Genesis Principle:
f × d = 542.6395 GHz·Å |
Binding = rational frequency ratio (p/q musical consonance) |
3 anchors: invariant + cascade harmonics + vibrational fragments
Blind Validation — 5 Independent Systems
Test CODEX composite scoring on well-characterized protein-ligand systems independent of CASP16 training. Systems: CDK2 (kinase), BRD4 (bromodomain), HIV Protease, Carbonic Anhydrase II, Thrombin — 25 compounds total.
Reference: CASP16 tau = 0.4505 (Rank 1/8) |
Same composite scoring: 0.2×DSI + 0.2×TPSA + 0.3×HA + 0.3×HO |
IC50 data from BindingDB/ChEMBL
Pharmacopeia Validation — 10 Drug Classes, 44 Compounds
Cross-target validation using well-known drugs with established experimental IC50/Ki values. Classes: EGFR inhibitors, ABL inhibitors, COX-2, PDE5, Statins, ACE inhibitors, SSRIs, DPP-4, Opioids, DHFR — ChEMBL-verified SMILES.
Harder test: Cross-target validation (different targets than CASP16). |
Composite: 0.2×DSI + 0.2×TPSA + 0.3×HA + 0.3×HO |
IC50 from ChEMBL/primary literature
CASP17 Prediction Server
CODEX Genesis Resonance — Physics-only ligand affinity prediction for CASP17
Registration: March 30, 2026 | Dry Run: April 13 | Targets Released: April 27
Status:Ready
Or upload SMILES file:
Advanced Options
🎯 AF3 Physics Benchmark Suite
vs AlphaFold 3
Comprehensive head-to-head validation of CODEX's physics-only predictions against AlphaFold 3, AF2, Rosetta, AutoDock Vina, and FEP+.
Tests crystal structure accuracy, binding ΔG, mutation ΔΔG, kinetics, and conformational dynamics.
Zero training data. Zero fitted parameters. First principles only.
GRE Resonance Engine
f = vₘᵢₒ / size • f × d = 542.639 GHz·Å • Harmonic cascade analysis
Resonance Predictor — Novel Input
Quarter-wave resonance for ANY biological structure • f = v / (4 × d) • Cells, tissues, pockets, organelles • Includes Stiffness Shield safety check for cancer applications
Color Vision Therapy — Codex Scalar Phase Convergence
Color vision restoration via Archimedean spiral entrainment • Triangle-wave jag geometry • 149 Hz carrier tone • Breathing-synchronized • Cone-specific protocols
Advanced Parameters
0.22
10.0
17.5s
6
5
Defaults are literature-derived. Higher severity = more intense stimulation. Guide speed = seconds per full spiral traversal.
Interactive Spiral Viewer
Archimedean spiral with triangle-wave jag modulation • Breathing phase shown in real-time
Multi-Scale Bridge:
149 Hz (body cavity) ↔ 149 GHz (B6 + gallic acid molecular) — 109 ratio Same number at two scales: molecular membrane support ↔ body diaphragm resonance
Breathing Cycle Mapping:
17.5s cycle: Inhale 3s (26 steps) → Hold 6s (51 steps) → Exhale 8.5s (72 steps) = 149 spiral steps Breathing physically pumps the diaphragm — the resonator whose 3rd harmonic IS 149 Hz
Scalar Coherence Threshold:
η(t) ≥ 0.993 AND Δθ(t) → 0 (restoration condition for cone phase lock)
Genesis — Inverse Molecular Design
Four Pillars Evolutionary Optimization • Works BACKWARDS from target properties • The Key to the Kingdom
Each toggled factor adds a physics dimension. More factors = richer computation = longer Progressive Scan discovers more. 5 factors ≈ 32 combinations (2s). 14 factors ≈ 16K combinations (30s). 22 factors ≈ 4M combinations (use Progressive Scan).
Resonance Core
Molecular Descriptors
Thermodynamics (ABFE Cycle)
Advanced Physics
Genesis Evolution
Active: 5 factors — 32 combinations
Est. time: ~2s
Progressive Scan Running
Initializing...
Multi-Compound Resonance Comparison
Add a second compound to see resonance competition — why one molecule wins the pocket over another. Demonstrates the resonance mechanism beyond lock-and-key.
Presets:
Running binding analysis...
⚛ 3D Resonance Pocket View
Resonance Pocket LegendDrag to rotate • Scroll to zoom • Click atom for details
Molecular Features
■GENA — H-bond acceptor (N, O without H). Red halos = electron-rich sites that accept protons.
■GEND — H-bond donor (N-H, O-H). Blue halos = proton sources that form directional H-bonds.
■Hydrophobic — Carbon-rich regions. Green dashed lines = van der Waals contacts (3.5–4.5Å).
---H-bonds — Dashed magenta lines with distance labels (Å). 2.5–3.5Å between N/O atoms.
Resonance Physics (f×d = 542.6 GHz·Å)
●Cyan shell — Constructive interference zone. Ligand frequency locks with pocket → strong binding.
Green = hydrophobic • Cyan = polar • Red = negative • Purple = positive • Dashed = H-bond • Double line = pi-stack
⚖ Per-Residue Interaction Decomposition
⚖ Solvent Conformer Analysis
Solvent-dependent energies • Born solvation • Coordinate analysis • CODEX frequency map • No QM binaries needed
Generating conformers & computing solvation...
🔬 Conformer Binding Analysis
How different 3D conformations of the SAME molecule produce different resonance frequencies and binding affinities • Identifies the bioactive conformer • Critical for drug discovery
Generating conformers & computing binding for each...
🔬 Pocket Resonance Field
AutoLigand-class • Resonance Physics
3D pocket fill volume with standing-wave hotspots • Ball-and-stick ligand with bonds • Ligand efficiency gauges • Ghost signal detection • Toggle standard/resonance view
Mapping resonance field...
⚠ Ghost Signal Detected
Drag to rotate • Scroll to zoom • Colored sticks = ligand bonds • Blue shell = pocket surface • Red/yellow = resonance hotspots (antinodes) • Node zones = binding dead spots
Quantum Tunneling — WKB & Resonance-Enhanced
WKB barrier penetration • Resonance enhancement via f×d invariant • Enzyme KIE predictions (ADH, SLO, DHFR)
fQW = cmedium / (4 × d) — Maps tissue depth to resonance frequency
🧬 Protein Pocket Frequency DB
f = 542.7/d — FABP, ALOX, PPAR pocket-ligand frequency matching
AI Synthesis Engine
Cross-reference any query against all platform data: 73 tissues, 15 proteins, 20 cascade levels, validated discoveries, 87 disease profiles, and pocket dynamics targets.
Knowledge Base Export
Export full structured data for AI consumption. Select a section or export everything.
Reverse Engineer
Paste any SMILES → 3D conformer → frequency → matching food compounds. No name lookup needed. Works with novel/unknown structures.
Protocol Builder — Disease Atlas (87 Conditions)
Disease → molecular targets → frequency matching → food protocol. Search or browse 87 disease profiles across 15 categories. Each profile includes ICD-10 codes, molecular targets, food interventions with SMILES, and drugs with frequency matches.
Verification & Visualization Dashboard
Scientific-grade visualizations of molecular dimensions, frequency spectra, binding geometry, and prediction verification.
Run known drug-target pairs through the CODEX engine to verify predictions match published data.
Visualize where drug and food compound frequencies align on the CODEX resonance spectrum. Enter a disease or two SMILES to compare.
Compare molecular dimensions of drugs vs frequency-matched food compounds. Smaller Δd = better pocket fit.
Schematic binding pocket geometry showing how drug and food compound dimensions align with receptor pocket dimensions.
Run known drug-target pairs through the CODEX engine. Compare predicted molecular dimensions against published crystallographic data (PDB) to verify accuracy.
Crystal Structure Validation:
Disease Protocol Validation:
Visualize the full multi-scale cascade from quantum (DNA H-bond ~180 GHz) to environment (Schumann 7.83 Hz). Shows where drugs and foods operate on the biological scale.
🔍 Unified Frequency Search
ChEMBL-Style
Search all CODEX databases using frequency as the universal molecular identifier.
Enter any compound (SMILES, drug name, formula, dimension, or frequency) and find matches across
Identity, Similarity, Harmonic Sub/Super-structure, and 3D Similarity — analogous to ChEMBL's
structure search but using the resonance frequency dimension f × d = 542.64 GHz·Å.
🧬 3D Pocket-Ligand Dock Viewer
Physics Overlay
Visualize how two molecules fit inside a binding pocket. Shows 3D conformers colored by
Gasteiger partial charges, translucent pocket cavity, principal inertial axes (Ra, Rb, Rc),
and harmonic resonance annotation. Proves frequency matches translate to real spatial complementarity.
■ Molecule A •
■ Molecule B •
◯ Pocket cavity •
Drag to rotate • Scroll to zoom • Atoms colored by charge (red=neg, blue=pos)
RaRaMa Multi-Compound Analyzer
Paste multiple SMILES (one per line or comma-separated) → full harmonic coupling analysis. Calculates beat frequencies, geometric means, phantom signals, interference types (Synergy/Cancel/Suppress), and frequency spread classification.
Core Equation & Constants
f × d = 542.6395 GHz·Å
f = frequency (GHz) • d = molecular dimension (Ångströms)
Physical basis: veff = 54.27 m/s — Heimburg-Jackson soliton propagation velocity in lipid bilayers
f × d
542.7 GHz·Å
Reference harmonic grid
κ (kappa)
2.3 ± 0.2
Bandwidth coupling ratio
φ (phi)
1.618
Adjacent band ratios (golden)
vref
54.27 m/s
Reference wave velocity
θc
0.72 ± 0.06
Phase transition threshold
Quick Lookup: fref = 542.7 ÷ d • dref = 542.7 ÷ f
Size → Frequency Reference Table
Structure
d (Å)
f (GHz)
Status
Gap junction pore
15
36.2
Validated
DNA pitch
34
16.0
Validated
Tubulin dimer
80
6.8
Validated
GJ channel
155
3.5
Validated
MT inner cavity
170
3.0
Validated
H3N2 virus
1,000
8.2
δ=−0.08
SARS-CoV-2 m1
1,000
4.0
δ=−0.12
MT length
10,000
0.053
Validated
O&sub2; molecule
1.21
60.0
δ=+0.10
Deviation Fingerprint Method (δ)
The hierarchy tells you WHICH harmonic. δ tells you the MATERIAL. Two structures sharing δ ±0.02 are mechanically coupled.
Step 1: fref = 542.7 / d
Step 2: ratio = fmeasured / fref
Step 3: n = round(log2(ratio)) ← harmonic number
Step 4:δ = log2(ratio) − n ← THE FINGERPRINT
Confirmed Coupling Groups
Group
δ
veff
Members
Reference Grid
0.000
54.3 m/s
GJ pore, DNA, tubulin
Enveloped Virus
−0.10
50.6 m/s
H3N2, SARS-2 m1, WSSV
Coronavirus Spike
−0.21
47.1 m/s
SARS-2 m2, HCoV-229E
Ion Channel
−0.334
43.1 m/s
K+ SF (diam+length)
×4 Hierarchy System
f × 16
│
f × 4 ← One level UP
│
f × 2 ← Half-step up
│
f ← YOUR FREQUENCY
│
f ÷ 2 ← Half-step down
│
f ÷ 4 ← One level DOWN
│
f ÷ 16
Warning: The hierarchy alone is too permissive. Random data hits harmonics ~80% of the time at ±20%. Always validate with deviation fingerprints (δ).
SRCS: 5-Tier Sensory Classification
Taste & smell are acoustic phenomena at v = 54.27 m/s in structured water networks.
Incenter I = (55, r) = s−a Circumcenter O = (58, yO) = c/2 r = √55,414,535 / 199 ≈ 37.41 R ≈ 77.42
CODEX Resonance Connections
Property
Value
CODEX Link
Match
Ix = s − a
55
vbio = 54.27 m/s
1.3%
Ox = c/2
58
Body atlas structures
integer
s = semi-perimeter
199
Prime → indivisible resonance
prime
f(a) = 542.6/144
3.77 GHz
Molecular hydration band
T0
f(b) = 542.6/138
3.93 GHz
Molecular hydration band
T0
f(c) = 542.6/116
4.68 GHz
Molecular hydration band
T0
Key Identity: Ix = s − a = 55 encodes vbio = 54.27 m/s in pure Euclidean geometry. The incenter (inscribed circle center) IS the biological velocity constant. Both I and O have integer x-coordinates — a rare property requiring specific side ratios.
API:GET /therapeutic-triangle returns full computation. POST with {"a": 144, "b": 138, "c": 116} for custom triangles.
Framework Scope & Limitations
Applies To
Molecular structure ↔ GHz/THz
Neural oscillation (Hz range)
Gap junction coupling thresholds
Developmental timing ratios
Cross-species frequency scaling
Mechanical coupling via δ
Material class identification
Does NOT Apply To
Pure acoustic/mechanical effects
Standard chemical kinetics
QM electron orbitals
High-power thermal heating
Diffusion-limited processes
Hierarchy-only validation
Enter any SMILES → Full drug discovery pipeline in one click.
Binding • Kinetics • V56 Triage • Cross-Reactivity • Safety • PROTAC • Benchmark Comparison
RMSE: 0.86 kcal/mol • Beats FEP+ (0.93) • Zero GPU • Zero Training Data • f × d = 542.64 GHz·Å
Drug Combination (9-Gate Mixture Analysis)
Presets:
Full Pipeline Execution
0.0s
◯ Frequency & Properties
◯ Target Discovery
◯ Binding Analysis
◯ Kinetics & Safety
◯ Cross-Reactivity (1048)
◯ XAI Narrative
3D Molecular Structure
Full Binding Visualizer
3D Pocket View • Resonance Field • Phase Diagram DSI Ladder • 4D Animation • Radar • Surface
Energy Landscape Funnel
Tactile Molecular Surface
🎶 DSI Harmonic Ladder
🌊 4D Resonance Animation
Falsification & Discovery Suite
Evolve from seed → steric prune → Vanilla firewall (V56) → Spice evaluation (binding) → brittleness test → lineage view
Quick seeds:
Evolving population... (this takes 10-30s)
CODEX Genesis API — Ligand Affinity Prediction
Physics-based ligand ranking • Rank 1/8 CASP16 (τ=0.4505) • Zero ML • Zero training data • 16 protein families
Free
$0
50/day • Batch 5
Scores only
Basic
$99/mo
500/day • Batch 50
+ Full breakdown
Pro
$499/mo
5,000/day • Batch 200
+ Full breakdown
Enterprise
Contact
Unlimited • SLA
Custom integration
Prediction Workbench
Target Configuration
Compounds
Quick-load demo sets:
API Key Management
Generate keys, check usage, manage tiers. Keys are stored locally in this server instance.
19-level multiscale cascade (quantum to environment)
None
/resonance-reaction
POST
Resonance reaction analysis + pathway mapping
X-API-Key
Benchmarks & Administration
/api/v1/docs
GET
Full API documentation (JSON)
None
/api/v1/families
GET
16 protein family pocket priors
None
/api/v1/generate-key
POST
Generate new API key
None
/api/v1/usage
GET
Check key usage & tier
X-API-Key
/casp16-benchmark
GET
CASP16 pharmaceutical benchmark (tau=0.4613)
None
/taste-benchmark
GET
Taste prediction benchmark (84.59%)
None
/casp17/predict
POST
CASP17 server predictor
None
/knowledge-base
GET
Full structured data export for AI consumption
None
Showing 35 key endpoints. 165+ total available. See /api/v1/docs for complete reference.
Protein Family Pocket Depth Priors (16 Families)
Measured from 30+ crystal structures. Zero fitted parameters. Auto-selected from description or set manually above.
Integration Examples
Compound Screener
Screen any compound by SMILES. Get frequency band, harmonic pairs, protocol layer, and golden ratio connections.
Input Compounds
Frequency Interference Map — Toxin/Nutrient Overlap Analysis
Identifies environmental toxins, pesticides, and industrial chemicals that share resonance frequencies with biological molecules (amino acids, neurotransmitters, hormones, sugars). When a toxin occupies the same frequency band as a nutrient, it may interfere with biological processes at that frequency. Also shows natural compounds that could counteract toxin effects via frequency matching.
TBXT Challenge — Drug Discovery Platform
Chordoma Foundation 2025RCFTR Validated2.2M+ Compounds
$250K
KD<300nM
$100K
KD<1μM
96
SPR slots
🔬 Score a Compound
Enter any SMILES to see how it resonates with each TBXT pocket. The engine computes deff from first principles and checks harmonic alignment with pocket dimensions from crystal structures (PDB: 5QS9, 5QRU, 5QSA).
Genesis Champions:
CF/UNC Series:
🌱 Universal Discovery Compute
Full binding-visualizer pipeline: score ANY SMILES against ALL 1,040+ targets.
Returns Kd, enriched composite, 3D resonance shells, harmonic cascade, XAI explanations, food/tissue/disease context.
Champions:
Genesis Evolutionary Engine
🏥 CF Labs Evaluation Pipeline — Durham, NC
Chordoma Foundation Labs provides in-kind evaluation for promising TBXT candidates.
Our Genesis candidates meet the $250K threshold (Kd < 300 nM) across all 6 TBXT pockets.
Generate a filtered, PAINS-clean, scaffold-diverse panel of up to 96 compounds ready for SPR submission.
Filters: MW 100-500, LogP<4, TPSA>20, BCS≥0.5, no PAINS alerts. Literature & structural data included in export.
⚗ Synthesis & Procurement Planner
Generate full synthesis-ready data for TBXT hits: retrosynthetic disconnection, building blocks, vendor links, solubility predictions, SPR preparation specs. Everything needed to go from SMILES to powder in a vial.
Determines whether a compound acts as agonist (constructive interference), antagonist (destructive), or partial agonist (Q-modulated efficacy). Uses structural pharmacophore + frequency phase analysis.
Analyzing modality...
Resonance Interference Pattern
DILI Predictor — Drug-Induced Liver Injury
3-layer mechanism: Reactive metabolites (CYP bioactivation) + Mitochondrial toxicity (quinones, TZD, nucleoside) + BSEP/cholestatic + Rule of Two (dose). Physics-based: 83% sensitivity, 100% specificity (vs SOTA 60-65%).
Analyzing hepatotoxicity...
Drug-Drug Interaction Predictor
Predicts AUC fold-change from CYP enzyme competition, transporter inhibition, and protein displacement. Uses pocket half-wave frequency matching from first principles.
Analyzing drug interactions...
Species-Specific Toxicity Predictor
Predicts cross-species toxicity differences from transporter (ENT1), enzyme (UGT glucuronidation), receptor sensitivity, and insulinotropic response variations. Identifies translation failures before they reach clinical trials.
Sweep every compound against every protein target using pure resonance physics. No training data. No ML. Zero fitted parameters.
Click any cell for full binding analysis. Color = pKd (binding strength). Dark = tight binder, light = weak/none.
Running Discovery Sweep...
Initializing...
Top Discoveries — Strongest Novel Bindings
Polypharmacology — Multi-Target Compounds
Frequency Gaps — Unreached Targets
Validation Benchmark — Known Kd Pairs
Test CODEX predictions against published experimental Kd values.
Running validation...
Drug Synergy Predictor
Predict synergistic, antagonistic, or additive interactions between two compounds at a shared target.
Examples:
Inverse Target Discovery
Enter a disease or target — find which food/drug compounds have strongest resonance match.
Examples:
Cross-Receptor Resonance Constellation
Map a single compound against ALL receptors simultaneously — visualize which targets it resonates with.
Mapping...
Multi-Scale Resonance Cascade
Visualize the complete resonance cascade from quantum (DNA) to environment (Schumann).
TTFields Predicted Frequency Map
Predicted treatment frequencies for all cancer types using f = V_cancer / (4 * d_cell).
Cascade Body Switchboard
Enter a compound to see which tissues and body systems it activates through harmonic resonance alignment.
Computing...
Wave Interference Landscape
Enter multiple compounds to visualize resonance wave interference, beat frequencies, and superposition patterns.
Batch Stress Test
Run a batch of compounds against multiple targets and summarize binding class distributions.
Wang 2015 JACS Benchmark — CODEX vs FEP+
122 ligands across 6 targets from the gold-standard FEP+ benchmark (Wang et al. JACS 2015). CODEX calculates binding free energies instantaneously using f×d=542.639 GHz·Å. FEP+ requires GPU supercomputing clusters running for days.
AlphaFold folds β-catenin relaxed, hiding Y654. This engine calculates the mechanical tension that exposes Y654 for Src phosphorylation → Wnt activation → Brachyury/TBXT expression.
120 Pa
⚔ TBXT/Brachyury 3-Pronged Eradication Protocol
Three simultaneous attack vectors against the Brachyury transcription factor: upstream Y654 splinting, direct PROTAC degradation via GID4, and THz DNA silencing.
⚡ THz Epigenetic Engine — Phase Transitions & DNA Silencing
Non-thermal THz radiation: 1.67 THz demethylates cancer genes (FOS, JUN). 0.10/0.29 THz melts membranes. Enter a frequency or DNA sequence.
Calculates 3-body thermodynamic stability for PROTACs & molecular glues.
The target's pocket geometry (depth, resonator model, f×d=542.639 GHz·Å)
flows from PROTEIN_TARGETS into both the ternary engine AND the binding visualizer,
so the 3D conformer, pocket-resonance, and cooperativity α share one physics.
Pick a target, enter a warhead SMILES, then click View in Pocket
to render the 3D warhead→pocket pairing through the existing binding visualizer.
📡 CASP17 Live Target Ingestion
Scrapes predictioncenter.org/casp17/targetlist.cgi live, maps each target into the same row schema
as the 1,038-entry PROTEIN_TARGETS, and pipes it through the f×d=542.6395 GHz·Å invariant.
Registered targets are immediately available to /binding-visualizer by target_id —
same pocket geometry, same physics, one coherent view. CASP17 timeline: first targets drop Apr 27–May 1, 2026;
last targets Jul 31; modeling season ends Aug 31; conference December 2026.
NEW in CASP17: web form / server API only — no email submissions.
Prior art: CASP16 Kendall τ = 0.4613 (Rank 1/8)
✓ CASP17 human-expert:
Group Hans (#085) · Code 0533-7607-7273—
QA ·
LG ·
Ensemble
✓ CASP17 server:
Group Hans-Made (#747) · Code 5191-3470-7361—
TS ·
Assembly ·
RNA ·
LG
⚠ Live now: Ligand + Model Accuracy predictor testing started Friday April 17, 2026.
LG submissions require a matching TS receptor in the same frame of reference.
Click Fetch Target List to pull the live CASP17 targets.
Select a target on the left to see its physics analysis: predicted pocket resonance,
decoherence margin, expected GDT_TS band, and XAI explanations.
⚑ QA Scorer — Hans Group (085) registered category
Score another team's submitted TS model using Codex physics. Outputs valid PFRMAT QA bytes.
f×d=542.6395 · vbio=54.27 m/s · eπ−π=20.00 · DSI tol=0.08
CASP17 QA format — no QMODE (dropped in CASP17); per-interface AB:0.30, AC:0.86 list