Ligand similarity-based virtual screening refers to find new ligands by evaluating the similarity between candidate ligands and the known active compounds. As one of the most powerful computer-aided drug design techniques, ligand similarity-based virtual screening relies heavily on several descriptors of molecular features, including atoms, the presence or absence of structural features, topological descriptors, geometry and volume, or stereoelectronic and stereodynamic properties.
Figure 1. Chemical Structure Similarity Search for Ligand-based Virtual Screening. (Xu, J.; et al. 2015)
Virtual screening based on the ligand similarity can be effectively applied to accelerate drug discovery. Our ligand similarity-based virtual screening can be divided into three categories:
1. Substructure screening: Molecular or physical and chemical global parameters, such as molecular volume, the number of hydrogen bond donors/acceptors, etc., are applied in substructure screening. We search for molecules in the database that contain query molecules as substructures.
2. Two-dimensional similarity screening: The properties of two-dimensional chemical structures are used to describe computational chemical similarities, such as molecular fingerprints (FP). FP is a typical path-based method that analyzes the linear path of all fragments of the molecular structure of a given number of bonds, which use simple vectors to characterize many chemical features. And the quantification of the similarity between FPs is usually obtained by the Tanimoto coefficient.
3. Three-dimensional similarity-WEGA: This method mainly considers the three-dimensional geometric conformation of molecules, including pharmacophore recognition and shape similarity. The most commonly used methods for three-dimensional similarity quantification are Tanimoto correlation coefficient and Manhattan distance. We mainly apply this method to study the skeleton migration, and use WEGA algorithm to calculate the three-dimensional structure similarity between query molecules and molecules in the database.
We have designed a multi-feature integration algorithm which is developed based on an algorithm-based matching and a machine learning containing various descriptors.
Our ligand-based is based on the molecular similarity evaluation between the submitted molecule(s) and those in an active compound database. The database is constituted by multiple reported bioactive molecules with target or mechanism information. We can conduct virtual screening and target prediction based on the two-dimensional and three-dimensional similarity evaluation of molecular structures.
Molecular structure similarity search is a common technique for drug discovery based on ligands. Its purpose is to identify candidate compounds that are similar in structure and biological activity to the query compound. Alfa Chemistry supports multiple descriptors of molecular features for high-throughput virtual screening:
The chemical fingerprint Tc is a common 3d similarity measurement used to calculate the fraction of the shared molecular volume between two ligands. We have developed a Rapid Overlap of Chemical Structure (ROCS) program based on the volume similarity.
Another 3D similarity measurement is pharmacological similarity, which only considers the volume overlap between key functional groups. We combine two-dimensional and three-dimensional indicators based on Obabel PF2 fingerprints, shapes, and pharmacological points for three-dimensional chemical similarity aggregation.
Our ligand similarity-based virtual screening services remarkably reduce the cost, promote further experiments, and accelerate the process of drug design for customers worldwide. Our personalized and all-around services will satisfy your innovative study demands. If you are interested in our services, please don't hesitate to contact us. We are glad to cooperate with you and witness your success!