Nine neurodegeneration-relevant proteins spanning Alzheimer's, Parkinson's, ALS, and Huntington's disease. The panel includes validated druggable targets (positive controls), known intractable targets (negative controls), and targets where the field remains uncertain. All results produced by the Ashebo engine from first-principles physics — no external database lookups.
| METRIC | ONCOLOGY | NEURODEGENERATION |
|---|---|---|
| Targets analyzed | 9 | 10 |
| Pursue | 2 (22%) | 5–6 (50–60%) |
| No-Go | 4 (44%) | 2 (20%) |
| Borderline | 1 (11%) | 2 (20%) |
| Median score | 31 | 76 |
The neurodegeneration panel has a higher pursue rate because the field has historically focused on enzyme targets (kinases, proteases, esterases) with well-defined active sites. The oncology panel was deliberately enriched for transcription factors and PPIs, which are structurally harder. This cross-indication comparison demonstrates consistent discrimination power while reflecting the genuine structural landscape of each field.
BACE1 and PSEN1 both score as druggable (72 and 83/100) despite their clinical failures. The engine correctly separates the structural question ("is there a pocket?") from the clinical question ("will binding it help patients?"). BACE1 inhibitors failed because of the amyloid hypothesis, not pocket quality. PSEN1 inhibitors failed because of Notch toxicity, not binding affinity.
Alpha-synuclein (0/100) and Tau (3/100) receive unambiguous no-go signals. These intrinsically disordered proteins cannot be addressed with conventional small-molecule binding. The engine provides clear discrimination that could save years of misguided small-molecule campaigns, redirecting resources toward immunotherapy or antisense approaches.
TDP-43 demonstrates domain-level discrimination: the RRM1 domain scores 76/100 as druggable, but the disease-relevant C-terminal low-complexity domain is intrinsically disordered. The engine identifies "this domain, not that one" — a distinction critical for therapeutic strategy in ALS and FTD.
The 312 Da C₁₀H₁₇N₂O₄ formula appears across LRRK2, BACE1, GSK3β, AChE, and PSEN1 — the same molecular weight optimum observed in the oncology sprint for KEAP1, RB1, and STAT3. This cross-indication convergence reinforces the hypothesis of a universal drug-like scaffold optimum for medium-sized enzyme active sites.