Atom Molecular Interaction Prediction

AI tool for the future precision medicine and precision health.

Molecular Interaction

Model trained on protein Delta Delta G change dataset. Predicts the different types of energy to calculate molecular interaction force.

In: Complex PDB file / Out: Energy Score

Image

Protein Force Field

Based on our different environment parameter to simulate molecular atom move and shake, Graphen Atom Protein Force Field Tool searches the structures with minimize energy or stable position. It relaxes protein side chains through torsion angle.

In: PDB file / Out: PDB file

Image

Protein-Protein Co-Simulation

Based on two PDB structures to simulate any possible contacting sites by torsion angle and coordinate distance, Graph-equivalent models use equivalent attention to learn atom features, position encoding, distance correlations and central tourism angles. Randomly collision energy prediction simulates more possible complexes by estimating contact energy. Models are trained on Protein-Protein and Antibody-Antigen complex data.

In: PDB structure file / Out: Complex PDB file

Image

Antibody-Antigen Contact Prediction

An epitope, also known as antigenic determinant, is the part of an antigen that is recognized by the immune system, specifically by antibodies B cells or T cells. According to statistics, 80% of epitopes will be conformational-form epitopes. It’s hard to predict based on sequence information. We use Graphen Atom Protein Structure Prediction Tool to search the surfaces with high affinity with BCR/TCR/MHC class. To Simulate the contact site between paratope and epitope, we wil need 10-25 length epitopes from the antigen structure.

In: PDB structure file / Out: Contacting face from antigen

Image

Speak to an expert today.