Protein modeling

Create accurate models of protein structures and understand binding mechanisms

Protein modeling is becoming an increasingly critical tool to fully understand chemical systems. Even when there is no crystal structure available for a target of interest, modeling methods can accurately propose likely 3D protein structures and binding mechanisms, which are essential for docking, virtual screening, and other structure‑based drug design methods.

Homology Modeling allows the creation of reliable 3D structures of your protein targets when no experimental structure is available. Protein homology models are valuable for finding potential pockets, grooves and binding sites for drug design, nucleic acid binding and protein-protein interactions: they help predict the bioactive conformation of a ligand, facilitating efficient ligand design.

The ever-growing number of available protein structures makes homology modeling a more feasible computational technique for protein structure prediction. Creating a homology model requires a correct primary sequence and 3D template protein structure(s) with high resolution and structurally complete, particularly in the active site or specific region of interest. An accurate homology requires that the templates share a significant sequence homology (usually a minimum of 35%) with the protein of interest. Once aligned, the differing amino acids in the template are replaced with the equivalent amino acids in the protein to model.

A known 4-Hydroxyphenylpyruvate dioxygenase (HPPD) X-ray structure (PDBID: 7CQS) (Left) used as template to create a homology model of the HPPD from Ambrosia artemisiifolia in a homology modeling experiment (Right).

The overall quality of a homology model can be assessed by Ramachandran plots, that can verify if the residue dihedral angles fall within permitted space, and protein contact map plots.

Figure 2. Ramachandran plot for a modeled protein illustrates whether the residue backbone dihedral angles, phi φ and psi Ψ, fall within permitted regions for ß-sheet, right-handed α-helix and left-handed α-helix.

The use of an ensemble of protein structures for modeling, the checking structural differences, and performing a multiple sequence alignment may help improve the quality of the model.

Placing structural or catalytic waters within the structure, modeling alternate loop conformations with the FREAD algorithm, performing molecular dynamics simulations and ensuring that the model represents the lowest energy structure can also improve the quality of the model.

Cresset Discovery Services has extensive experience in homology modeling and can support/work alongside you on your project.