VectaBind combines SE(3)-equivariant geometry with ESM2-3B protein language model embeddings to score binding affinity in under a second. Upload a compound library, triage HTS hits by indication, or screen selectivity across a target panel — all in the browser or via REST API.
500 free scores/month · No credit card · Get free API key
VectaBind's 0.20 pKd MAE was measured on the PDBBind 2020 validation split, which is used for model selection and tends to be optimistic relative to a held-out test set. The other models' numbers come from their published papers and may use different splits — these bars are reference points, not a head-to-head comparison.
1 pKd unit = 10× difference in binding affinity. Experimental reproducibility floor across labs ~0.40 pKd. Bars are not a controlled comparison — different splits, different evaluation protocols.
At sub-0.21 pKd MAE on the PDBBind 2020 validation split, VectaBind is approaching the reproducibility limits of wet lab assays across different labs. This is a model-selection number — a controlled evaluation on a held-out test set (CASF or similar) is in progress.
Submit compound SMILES strings and a target protein ID. Get binding affinity predictions back in under a second.
Import CSV with SMILES, compound IDs, and series tags. Validate structures before scoring — no ELN integration required.
Filter targets by disease area — AML, breast, lung — instead of hunting PDB codes. Each target shows gene symbol and structure ID.
Rank against one target or run a multi-target selectivity heatmap. Sortable tables, pKd vs LogP SAR view, export CSV for your team.
Design analogs with REINVENT4, push top hits back into your library, and re-score with the full affinity model.
Interactive walkthrough: Hit Triage Workbench guide · Launch app
Pre-computed pocket embeddings for clinically relevant kinases, GPCRs, and enzymes. The free tier includes core disease panels; Pro unlocks the full 473-target catalog across 19 therapeutic areas.
EGFR, KRAS, CDK4/6, BRAF, HER2, VEGFR2, MET, ALK + 70 more
BACE1, MAPT (tau), SNCA (α-syn), LRRK2, APP + 30 more
ACC2, PCSK9, Factor Xa, Thrombin, ACE2 + 40 more
JAK1/2, TNF-α, IL-6R, COX-2, PDE4, BTK + 45 more
SARS-CoV-2 Mpro, Influenza NA, HIV-1 PR, TB InhA + 35 more
D2R, 5-HT2A, SERT, NET, GABA-A, MAO-A/B + 30 more
CFTR, SMN1, dystrophin, PAH, GBA + 25 more
GLP-1R, PPAR-γ, DPP-4, SGLT2, thyroid receptors + 35 more
Liver, lung, bone, skin, eye, kidney, reproductive, aging, natural medicine
Three components that work together to exceed what any single approach can achieve.
Processes 3D pocket geometry using equivariant graph neural networks. Predictions are invariant to protein orientation — a fundamental physical constraint previous models had to learn from data.
8 layers · k=8 neighbors · sparse message passing3-billion parameter protein language model pretrained on 250M sequences provides rich evolutionary and functional context. Captures binding site properties that coordinates alone cannot encode.
2560-dim embeddings · per-residue contextBidirectional cross-attention models complex ligand-pocket interactions across 6 blocks and 12 heads. Each head learns distinct interaction patterns — hydrophobic contacts, H-bonds, electrostatics.
6 blocks · 12 heads · 65M total paramsThe VectaBind app gives you a complete environment: 3D protein visualization, compound scoring, AI molecule generation with REINVENT4, and an AI assistant powered by Claude — all in your browser.
No credit card required for the free tier. Upgrade when you need more throughput or targets.
Get free API access in minutes. Score your library on 70+ targets today — upgrade to Pro for the full 473-target catalog.
Get free API key →Questions? [email protected] · Response within 24 hours