GENERATIVE COMPUTATIONAL BIOLOGY

Precision Drug Design
at Atomic Scale

Accelerate molecular discovery with ultra-fast structure prediction, target docking, and automated conformer intelligence.

45savg run time
100%provenance
Zerolocal setup
Capabilities

Engineered for real science

Every tool in the pipeline exposes raw data, provenance, and deterministic parameters.

Automated Pocket Detection & Cavity Topology

fpocket integration with Voronoi-based druggability scoring

Detected Pocketssample data
PocketVolumeDruggabilityHydrophobic
#11,240 ų0.8462%
#2890 ų0.6148%
#3420 ų0.3129%
Center [X: 12.4, Y: -8.1, Z: 34.2]|Radius: 9.0 Å

Distributed Vina Molecular Docking

Cloud worker pool with exhaustiveness scaling

vina-dock.sh
vina --receptor target.pdbqt \
     --ligand ligand.pdbqt \
     --exhaustiveness 32

Mode | Affinity (kcal/mol) | RMSD
   1  |           -9.4      | 0.000
   2  |           -8.7      | 1.412
   3  |           -8.1      | 2.105

One-Click Structure Refinement

Automated preparation with OpenBabel integration

Auto-assign Gasteiger charges
Strip non-crystallographic water
Optimize flexible rotatable bonds
Torsions: 6/6 active
Add polar hydrogen donors

Cryptographic Provenance & Reproducibility

Every result carries an auditable, seed-locked execution manifest

provenance-manifest.json
{
  "seed": 14920481,
  "engine": "vina-v1.2.5",
  "box_size": [20, 20, 20],
  "center": [12.4, -8.1, 34.2],
  "exhaustiveness": 32,
  "provenance": "verified"
}
Sample manifestSHA-256: a4f2...e81c
Workflow

From sequence to insight

A continuous, auditable pipeline — no manual handoffs.

01

Ingestion

Direct upload from RCSB PDB ID or AlphaFold DB.

02

Pocket Search

Instant Voronoi cavity partitioning via fpocket.

03

Batch Docking

Cloud worker pool execution with AutoDock Vina.

04

Pose Scoring

Energy-minimized pose ranking and CSV/PDBQT export.

Active Runexample preview
7K40 → CHEMBL25·exhaustiveness=32
42.3s execution time
Mode 1: -9.4 kcal/mol72% complete · 23/32 modes
Benchmarks

Built for scientific rigour

Illustrative target metrics. No hardware or load measurements have been taken yet.

< 45s

Single-run average docking latency

10,000+

Ligands per batch virtual screening

100%

Deterministic reproducibility with seed lock

Zero

Local CLI dependencies or drivers needed

Developer API

Trigger runs from anywhere

Full Python SDK, REST API, or point-and-click UI. Same deterministic engine underneath.

import ligentra

client = ligentra.Client(api_key="lg_live_...")
job = client.docking.create(
    protein="7K40.pdb",
    ligand="CHEMBL25.sdf",
    engine="autodock-vina",
    exhaustiveness=32
)
print(f"Top Pose: {job.best_affinity} kcal/mol")

Start with 100 free credits

No credit card required. Full access to the docking engine, structure prediction, and provenance tools.