GPC3 hepatocellular carcinoma target · five-stage digital discovery
From one sequence to one
antibody candidate.
Built on real structural data, this interactive computational workflow shows how AI identifies a target, predicts epitopes, generates antibodies, and completes virtual validation.
TARGETGPC3P51654 · 580 aa
Target sequence580 aa● Real data
AlphaFold global pLDDT75.08● Real data
Candidate epitopes3◇ Simulated data
Antibody candidates4◇ Simulated data
END-TO-END WORKFLOW
0/5 stages completedFive-stage discovery pipeline
01READY
DIGITAL ANTIGENDigital antigen construction
Build an atomic 3D model from the amino-acid sequence
OUTPUTAntigen structure · PAE · confidence
↗02QUEUED
EPITOPE PREDICTIONAI epitope prediction
Scan surface accessibility and conserved regions
OUTPUTTop Epitope · site scores
↗03QUEUED
ANTIBODY GENERATIONAI antibody generation
Generate complementary antibody sequences from the epitope
OUTPUTVH/VL · CDR · Fab/VHH
↗04QUEUED
VIRTUAL SCREENINGVirtual screening
Evaluate candidate developability across multiple parameters
OUTPUTComposite score · candidate ranking
↗05QUEUED
BINDING SIMULATIONBinding simulation
Resolve complex poses and interface interactions
OUTPUTComplex · ΔG · KD
↗