Validation & evidence

Lead with reproducible outputs you can verify — scored libraries, ChEMBL analogs, RDKit properties, and CSV exports. Absolute affinity on every marketed drug is harder than rank-ordering within your own series; both views are documented here.

Reproduce in the app: Open Pathfinder demo for a pre-scored 10-compound EGFR library, or paste any SMILES below into the workbench. Scores generated with VectaBind API v1.0.0 · EGFR pocket (PDB 2ITX).

EGFR scored library (reproducible export)

Real batch output from the Hit Triage Workbench — 10 generated EGFR compounds ranked by predicted pKd. Open the demo to inspect 3D pocket overlay, MPO filters, and CSV export.

Rank	ID	pKd	QED	LogP	Lipinski	ChEMBL top sim	Confidence
1	GEN-16	8.18	0.92	1.82	yes	—	high
2	GEN-11	8.16	0.94	2.95	yes	—	high
3	GEN-18	8.09	0.92	1.97	yes	61%	high
4	GEN-20	8.09	0.92	2.37	yes	—	high
5	GEN-13	8.07	0.94	2.26	yes	—	high
6	GEN-12	8.06	0.94	3.06	yes	80%	high
7	GEN-15	7.86	0.94	2.44	yes	—	high
8	GEN-14	7.85	0.95	2.52	yes	—	high
9	GEN-19	7.80	0.94	3.85	yes	—	high
10	GEN-17	7.67	0.94	3.15	yes	—	medium

GEN-12 ChEMBL top hit: CHEMBL551595 (80% Tanimoto). Export includes all columns for team review.

Spotlight: GEN-12 · EGFR hit #6

Example of orthogonal signals med chem teams expect — not a single opaque score.

EGFR · rank #6 Strong binder

GEN-12 · Gen-EGFR

Predicted pKd8.06

QED0.94

ChEMBL similarity80%

LipinskiCompliant

LogP3.06

ConfidenceHigh

Drug-like · MW 317 · bind 76.5% · verify analogs in ChEMBL yourself

External spot-check: known EGFR inhibitors

Literature pKd values from ChEMBL bioactivities (kinase assays; rounded). VectaBind predictions scored live against EGFR pocket 2ITX, June 2026. Assay type and pocket representation differ from PDBBind training — use for transparency, not as a marketing MAE claim.

Compound	ChEMBL	Literature pKd	VectaBind pKd	\|Δ\|
Erlotinib	CHEMBL553	9.00	5.74	3.26
Gefitinib	CHEMBL939	9.00	5.65	3.35
Lapatinib	CHEMBL554	8.52	7.29	1.23
Osimertinib	CHEMBL3353410	9.70	7.42	2.28
Afatinib	CHEMBL1173655	9.30	7.96	1.34
Vandetanib	CHEMBL941	7.40	7.47	0.07
Neratinib	CHEMBL209386	8.70	5.56	3.14
Dacomitinib	CHEMBL3989959	8.82	7.54	1.28
Canertinib	CHEMBL3035	8.43	5.79	2.64

How to read this: VectaBind is optimized for rank-ordering compounds within your library on a fixed target. Absolute affinity on approved quinazoline drugs can diverge from cellular/biochemical literature values. Three of nine spot-check compounds fall within 1.0 pKd — Vandetanib within 0.1. Re-score any SMILES in the app to verify.

Signal provenance

Every column in the workbench traces to a named source — not a black-box number.

Signal	Source
pKd / bind probability	VectaBind Stage 6 EGNN + calibration
QED, MW, LogP, Lipinski	RDKit (in API + app)
ChEMBL similarity	ChEMBL REST API (browser-side lookup)
MPO composite	Weighted pKd + QED + Lipinski (app workflow)
3D pocket	PDB structures · EGFR 2ITX
CSV export	Full scored row from workbench session

Model benchmark (secondary)

Internal PDBBind 2020 validation MAE is reported in Methods. For screening workflows, rank order within your library and ChEMBL-grounded analog checks are more actionable than a single global MAE.

Metric	Value	Notes
PDBBind 2020 val MAE	0.20 pKd	Model-selection split · see Methods
Intended use	Relative ranking	Compare compounds on the same target in your library