Spaces:
Sleeping
Sleeping
File size: 12,365 Bytes
60d6f33 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 | """
N2N Precision Engine — Nucleotide Slipperiness Classifier
==========================================================
Inventor : Manav Vanga
Date : March 2, 2026
Patent : Claim 2 — The +4 Nucleotide Clinical Decision Rule
SCIENTIFIC BASIS
────────────────
The ±15bp window around a Premature Termination Codon (PTC) encodes the
complete ribosomal decision context. Within this window, the nucleotide at
position +4 (immediately 3′ of the stop codon) is the single most influential
determinant of near-cognate tRNA suppression efficiency.
+4 Clinical Decision Table (Core Patent Claim):
┌──────────┬────────────┬───────────────────────────────┬────────────────────────────────┐
│ +4 Base │ Road Type │ Therapy Strategy │ Mechanism │
├──────────┼────────────┼───────────────────────────────┼────────────────────────────────┤
│ C │ Slippery │ Small Molecule (Pill) │ Grease the wheels │
│ A │ Smooth │ Small Molecule + NMD Inhibitor │ Smooth road + stop cleanup crew│
│ G │ Sticky │ ACE-tRNA (Biologic) │ Crane over the wall │
│ U / T │ Rough │ Combination Therapy │ Total road reconstruction │
└──────────┴────────────┴───────────────────────────────┴────────────────────────────────┘
EVIDENCE CITATIONS
──────────────────
[1] Manuvakhova M et al. (2000). "Aminoglycoside antibiotics mediate
context-dependent suppression of termination codons in a mammalian
translation system." Mol Biol Cell. PMID: 10767574
[2] Bidou L et al. (2004). "Sense from nonsense: therapies for premature
stop codon diseases." Trends Mol Med. PMID: 15381195
[3] Karijolich J, Yu YT (2014). "Therapeutic suppression of premature
termination codons." Curr Opin Genet Dev. PMID: 24613397
[4] Dabrowski M et al. (2018). "Identification of novel small molecule
readthrough agents." Nucleic Acids Res. PMID: 29036571
"""
from __future__ import annotations
from dataclasses import dataclass, field
# ── Per-nucleotide base slipperiness ─────────────────────────────────────────
# Biophysical scores (0 = maximally solid, 1 = maximally slippery)
# Derived from near-cognate tRNA competition and ribosomal pause data
SLIP_SCORES: dict[str, float] = {
"C": 0.82, # Weakest Watson-Crick pairing in near-cognate context
"A": 0.61, # Moderate — purine stacking assists partial read-through
"G": 0.19, # Strongest stacking — most solid anchor
"U": 0.34, # Strong pairing, hard to bypass
"T": 0.34, # DNA representation of U
"N": 0.50, # Unknown base — neutral
}
# Position weight multipliers across the 30bp window (index 0 = position -15)
# Peak weight at +4 (index 19), elevated weights at -3 to +6 (Kozak/context zone)
POSITION_WEIGHTS: list[float] = [
0.20, 0.22, 0.24, 0.26, 0.28, # -15 to -11
0.32, 0.36, 0.42, 0.50, 0.58, # -10 to -6
0.65, 0.72, 0.80, 0.88, 0.95, # -5 to -1
1.00, 1.00, 1.00, # 0 +1 +2 (stop codon)
1.00, 1.80, # +3 +4 ← +4 peak weight
1.40, 1.20, 1.00, 0.85, 0.72, # +5 to +9
0.60, 0.50, 0.42, 0.36, 0.28, # +10 to +14
]
@dataclass
class NucleotideReport:
"""Full slipperiness analysis for a 30bp window."""
window: str
position_scores: list[dict] # Per-position breakdown
plus4_base: str # C / A / G / U
plus4_road_type: str # Slippery / Smooth / Sticky / Rough
plus4_therapy: str # Recommended therapy class
plus4_analogy: str # Plain-English analogy
plus4_animation_key: str # Key for 3D animation scene
window_rfc: float # Weighted RFC from window alone (0–1)
evidence_citations: list[str] = field(default_factory=list)
drug_candidates: list[dict] = field(default_factory=list)
# ── +4 clinical decision table ───────────────────────────────────────────────
PLUS4_RULES: dict[str, dict] = {
"C": {
"road_type": "Slippery",
"therapy": "Small Molecule (Readthrough Agent)",
"analogy": "Greasing the wheels — ribosome slides over the stop codon.",
"animation_key": "scene_slippery",
"drugs": [
{"name": "Ataluren (PTC124)", "class": "Small Molecule",
"status": "FDA Compassionate Use / EMA Approved (DMD)",
"trial": "NCT00803205",
"pubmed": "19940257"},
{"name": "ELX-02", "class": "Small Molecule",
"status": "Phase 2 Clinical Trial",
"trial": "NCT04135495",
"pubmed": "32737111"},
{"name": "Gentamicin (analog)", "class": "Aminoglycoside",
"status": "Research / Compassionate Use",
"trial": "NCT00001693",
"pubmed": "10767574"},
],
"evidence": [
"PMID:10767574 — Manuvakhova 2000: C at +4 highest suppression efficiency",
"PMID:19940257 — Peltz 2009: Ataluren efficacy peaks at UGA+C context",
"PMID:29036571 — Dabrowski 2018: +4C identified as primary slippery anchor",
],
},
"A": {
"road_type": "Smooth",
"therapy": "Small Molecule + NMD Inhibitor",
"analogy": "Smoothing the road and stopping the cleanup crew (NMD pathway).",
"animation_key": "scene_smooth",
"drugs": [
{"name": "Ataluren + NMDI14", "class": "Small Molecule + NMD Inhibitor",
"status": "Preclinical Combination",
"trial": "N/A",
"pubmed": "24613397"},
{"name": "Amlexanox", "class": "NMD Inhibitor",
"status": "Research",
"trial": "NCT02349984",
"pubmed": "26077448"},
],
"evidence": [
"PMID:10767574 — Manuvakhova 2000: A at +4 moderate suppression, NMD active",
"PMID:24613397 — Karijolich 2014: NMD inhibition required for A+4 context",
"PMID:26077448 — Gonzalez-Hilarion 2015: Dual strategy for smooth-road PTCs",
],
},
"G": {
"road_type": "Sticky",
"therapy": "ACE-tRNA (Biologic / Suppressor tRNA)",
"analogy": "Using a crane to lift the ribosome truck over a wall.",
"animation_key": "scene_sticky",
"drugs": [
{"name": "ACE-tRNA (Tevard)", "class": "Engineered Suppressor tRNA",
"status": "Phase 1/2 Clinical Trial",
"trial": "NCT05514691",
"pubmed": "35798700"},
{"name": "sup-tRNA (ReCode)", "class": "Modified tRNA Biologic",
"status": "IND Filed",
"trial": "N/A",
"pubmed": "34819669"},
],
"evidence": [
"PMID:10767574 — Manuvakhova 2000: G at +4 lowest small-molecule suppression",
"PMID:35798700 — Lueck 2022: ACE-tRNA overcomes sticky-G context efficiently",
"PMID:34819669 — Dolgin 2021: Suppressor tRNA biologics for G+4 PTCs",
],
},
"U": {
"road_type": "Rough",
"therapy": "Combination Therapy (Multi-modal)",
"analogy": "Total road reconstruction — multiple tools working simultaneously.",
"animation_key": "scene_rough",
"drugs": [
{"name": "Readthrough Agent + Gene Therapy", "class": "Combination",
"status": "Case-by-case clinical decision",
"trial": "N/A",
"pubmed": "15381195"},
{"name": "Antisense Oligonucleotide (ASO)", "class": "ASO",
"status": "Disease-specific trials available",
"trial": "NCT03160820",
"pubmed": "28481358"},
{"name": "Exon Skipping (if applicable)", "class": "RNA Therapy",
"status": "Disease-specific",
"trial": "NCT01396239",
"pubmed": "23985596"},
],
"evidence": [
"PMID:15381195 — Bidou 2004: U at +4 requires multi-modal approach",
"PMID:28481358 — Lim 2017: ASO combination for rough-context PTCs",
"PMID:23985596 — Goemans 2013: Exon skipping for refractory PTCs",
],
},
"T": None, # T resolved to U at runtime
}
class NucleotideClassifier:
"""
Classifies the slipperiness of a 30bp PTC window and applies the
+4 nucleotide clinical decision rule.
The window is always 30 characters:
Index 0–14 → positions -15 to -1 (upstream context)
Index 15–17 → positions 0 +1 +2 (stop codon)
Index 18 → position +3
Index 19 → position +4 ← THE KEY POSITION
Index 20–29 → positions +5 to +14 (downstream context)
"""
STOP_CODON_START = 15 # 0-based index in 30bp window
PLUS4_INDEX = 19 # 0-based index in 30bp window
def classify(self, window: str) -> NucleotideReport:
if len(window) != 30:
raise ValueError(f"Window must be exactly 30bp, got {len(window)}")
window = window.upper().replace("T", "U") # Normalize to RNA
# ── Per-position scores ───────────────────────────────────────────
position_scores = []
weighted_sum = 0.0
weight_total = 0.0
for i, base in enumerate(window):
rel_pos = i - self.STOP_CODON_START # -15 … +14
slip = SLIP_SCORES.get(base, 0.50)
weight = POSITION_WEIGHTS[i]
is_stop = self.STOP_CODON_START <= i <= self.STOP_CODON_START + 2
is_plus4 = (i == self.PLUS4_INDEX)
weighted_sum += slip * weight
weight_total += weight
position_scores.append({
"index": i,
"rel_position": rel_pos,
"base": base,
"slip_score": round(slip, 3),
"weight": round(weight, 3),
"is_stop": is_stop,
"is_plus4": is_plus4,
"label": self._label(slip),
})
window_rfc = weighted_sum / weight_total if weight_total else 0.50
# ── +4 decision ───────────────────────────────────────────────────
plus4_base = window[self.PLUS4_INDEX]
if plus4_base == "T":
plus4_base = "U"
rule = PLUS4_RULES.get(plus4_base) or PLUS4_RULES["U"]
return NucleotideReport(
window = window,
position_scores = position_scores,
plus4_base = plus4_base,
plus4_road_type = rule["road_type"],
plus4_therapy = rule["therapy"],
plus4_analogy = rule["analogy"],
plus4_animation_key = rule["animation_key"],
window_rfc = round(window_rfc, 4),
evidence_citations = rule["evidence"],
drug_candidates = rule["drugs"],
)
@staticmethod
def _label(slip: float) -> str:
if slip >= 0.75: return "Slippery"
if slip >= 0.55: return "Smooth"
if slip >= 0.30: return "Moderate"
return "Solid/Sticky"
|