LGNM Learned Graph Network Model

Predict per-residue flexibility, normal modes, and conformational dynamics using a Learned Graph Network Model with quantum-inspired neural PDE operators.

Analysis Suite

Residue-Level Flexibility Prediction

Predict residue-wise flexibility directly from protein structures using a learned Graph Network Model integrated with graph neural networks, providing high-resolution dynamic profiles without molecular dynamics simulations.

Collective Motion Analysis

Compute the lowest-frequency normal modes to characterize intrinsic collective motions associated with conformational changes, domain rearrangements, and functional dynamics.

Quantum-Inspired Computational Engine

Employ a quantum-inspired Hamiltonian formulation to efficiently solve network dynamics, reducing computational complexity while maintaining accurate flexibility estimation for large proteins.

Interactive Structural Visualization

Explore predicted flexibility through an interactive three-dimensional molecular viewer with residue coloring, structural annotations, measurements, and publication-ready rendering.

Comprehensive Analytical Reports

Generate publication-quality RMSF profiles, flexibility distributions, normal mode visualizations, residue statistics, and structural summaries for downstream interpretation.

Exportable Research Outputs

Download annotated protein structures, residue-level prediction tables, processed datasets, high-resolution figures, and analysis reports in standard research formats.

Analysis Workflow

1

Load Protein Structure

Enter a Protein Data Bank (PDB) accession (e.g., 6LU7) or upload your own .pdb or .cif structure for analysis.

2

Configure Analysis

Select the protein chain(s) to analyse and configure the prediction settings before initiating the computation.

3

Explore and Export Results

Visualize residue-level flexibility in an interactive 3D viewer, examine analytical plots, and download all generated structures, figures, and result files.