A well-designed compound library contains a broad range of truly stored chemicals and/or virtual compounds, which can meet majority requirements and truly jump-start the search for new drugs. In addition, such compound library or chemical library is able to provide comprehensive associated data of the stored chemicals, including chemical structure, purity, quantity, and physicochemical properties. Generally, a compound library is mainly composed of a series of 2D or 3D representations of compounds, which can be applied for various purpose through computational approaches. In order to increase the screening success rate and reduce unnecessary costs, different screening strategies have been developed and widely used. According to the design purpose, the screening strategies can be roughly divided into two types: positive design and negative design.
Figure1. Negative design and positive design screening concept diagram(Yang, Z.Y.; et al. 2020)
Positive design is mainly to form compound libraries through narrowing the scope of compounds and focusing on potential candidates with desired properties and structures. Generally, two methods based on receptor structure and ligand structure are used for design.
Receptor refers to the specific macromolecular structure on the cell membrane or in the cell of an organism, including enzymes, ion channels, antigens, nucleic acids, carbohydrate macromolecules, lipids, etc. The receptor-based design method starts from the structure and properties of the receptor to find ligand molecules that can specifically bind to it, including specific methods such as molecular docking based on the receptor structure, active site analysis, and de novo design.
Ligands refer to biologically active substances that can specifically bind to receptors, including information molecules and drugs. The ligand-based design method constructs a structure-activity relationship or pharmacophore model based on a leader molecule with known activity, including quantitative structure-activity relationship (QSAR), pharmacophore model, receptor mapping, molecular shape-based superposition and other specific methods.
Negative design mainly studies how to effectively delete compounds or noise molecules that contain undesirable properties or structures in the compound library, so that the remaining data can show regularity according to the order predicted by the expected model. According to the screening purpose and specific properties, negative design can be divided into drug-likeness, frequent hitters and toxicity, which are respectively related to the potential, selectivity and safety of drug molecules. The most famous negative design today is Lipinski's Rule of Five.
The concept of drug-like drugs is to explore the relationship between molecular characteristics such as structural characteristics and physicochemical characteristics and molecular behavior in the body, which is mainly reflected in solubility, permeability, metabolic stability and transportation efficiency. The existing drug-like rules mainly screen the physical and chemical properties of drugs.
High-throughput screening methods are widely used in drug development. However, a large number of compounds have been found to show positive results in different types of target screening. Such compounds are called "frequent hitters". There are many mechanisms for frequent hitters, mainly non-specific bond formation reactions or interference from experimental reagents. It should be noted that not all frequent hitters are undesirable.
Screening and eliminating potentially toxic compounds is one of the important ways to reduce the cost of drug development and increase the rate of drug development results. Considering the ease of interpretation and efficiency of the screening rules, the toxic structure screening rules have been widely used in drug screening.
BOC Sciences’ composite library integrated services are tailored for each project to ensure that goals are met or exceeded. We work with a wide range of organizations, from virtual companies to large pharmaceutical companies, including many of the world’s top companies, as well as academic research groups from around the world.