Structure-based drug design (SBDD) and ligand-based drug design (LBDD) are two basic methods of computer-aided drug design (CADD) applied in modern drug discovery and development. Virtual screening, as an important tool for high-throughput screening (HTS) to identify potentially bioactive compounds, has been widely used in drug development. Molecular docking can be used for virtual screening of new ligands based on biological structures and can be used to find targets and identify and optimize lead compounds in drug development. Therefore, molecular docking is a kind of SBDD and plays an important role in the rational design of new drug molecules.
The quantitative structure-activity relationship (QSAR) is an important chemometric tool in the computational design of drugs and common practice in LBDD. QSAR research provides structural features/physical related to structurally similar molecules and their biological activities Information on chemical properties.
Some people call compounds that meet the drug-like 5 rules (RO5) drug-like compounds. Drug-like is an indispensable intrinsic property of drugs, which plays an important role in the metabolism, pharmacokinetics, and toxicity of drugs in vivo. In the process of drug development, only by continuously evaluating the drug-like properties of drug molecules can high-quality drug candidates be screened. Virtual screening can evaluate the drug-like properties (ADME/toxicity) of potential drugs, which also greatly improves screening efficiency, reducing research and development costs.
Molecular docking is a technique used to determine the binding affinity between a protein and a ligand. It can simulate the process of molecular recognition by predicting the free energy of binding and the interaction of the binding protein-ligand complex. Therefore, molecular docking can predict the preferential orientation of ligand binding to a specific protein. Structure-activity relationship (SAR) is the study of the relationship between the chemical structure of a molecule and its biological activity. It is one of the most traditional drug research methods, which study the biological activity of a molecule to modify its structure.
Virtual Screening And Drug Similarity
Drug-like
In the preclinical and clinical development stages, candidate drugs often fail due to lack of efficacy or unrelated to the efficacy of the expected drug target. The main reason is poor drug-like properties, and pharmacokinetics and toxicity issues account for more than half of clinical trial failures.
TheThe RO5 rule is considered a guideline for predicting oral bioavailability, but 16 percent of oral drugs violated at least one of the criteria, and 6 percent failed in two or more. The RO5 rule is also often seen as a drug discovery filter that not only reduces the molecular library for virtual screening but also eliminates some ligands before synthesis.
Significance Of Drug Similarity Screening
Continuous evaluation of the physical and chemical properties affecting pharmacokinetics and toxicity, and optimization of the structure based on this, can improve the efficiency of drug development and reduce the cost. Therefore, in virtual screening, lead identification and lead optimization is an iterative process, starting with virtual prediction.
Bioactivity and drug-like properties are complementary, and both are necessary for a good drug. Compounds with good biological activity or selectivity may not be the best drugs, because drug-like properties may lead to poor drug metabolism and high toxicity, while compounds with poor biological activity but good drug-like properties may produce better clinical efficacy.
Molecular Docking
The main purpose of molecular docking is to enable both the target and the ligand to achieve an energetically favorable confirmation so that the ligand molecules are optically interconnected.
Docking Components
Molecular docking is mainly composed of search algorithm and scoring function. The search algorithm is used to determine the optimal conformation of the target-ligand complex, namely the position and orientation of the two molecules relative to each other. The resulting complex and the energy of each individual interaction can also be calculated. More clearly, the search algorithm can also generate “poses,” or the orientation of a particular conformation of a ligand at the active (binding) site of a protein molecule.
Docking Method
Molecular docking has two functions: search for conformation and ligand site binding. Therefore, molecular docking studies are often used to predict the binding orientation (binding mode/posture) of small molecules to their proteins (receptors) to predict the affinity and activity of small molecules.
Structure-activity Relationship (SAR)& Quantitative Structure-activity Relationship (QSAR) Research
SAR
The main chemical groups causing the biological reaction of drug molecules can be determined by SAR analysis, and their biological activity can be affected by modifying the chemical structure. Medicinal chemists use chemosynthesis and computational drug design techniques to insert new chemical groups or change existing ones into bioactive molecules and test for biological reactions that change (increase or decrease). SAR information is mainly obtained by experimentally determining that the biological activity of compounds varies with their structures. AR assumes that molecules with similar structures have similar activities, because similar compounds may have similar chemical or physical-chemical properties.
QSAR
QSAR is a mathematical relationship between chemical structure and biological activity. While the SAR hypothesis is based on the assumption that structurally similar molecules have similar functions (activity), QSAR assumes that the biological activity (similar activity) of different molecules can be quantitatively compared based on the characteristics of their structural components. The steps involved in QSAR analysis include data set preparation, structural optimization, calculation and selection of molecular descriptors, the establishment of related models, and final evaluation and validation.
QSAR has two types: 2D-QSAR and 3D-QSAR.
2D-QSAR is a drug design method that takes the overall structural properties of the molecule as a parameter, performs regression analysis on the physiological activity of the molecule, and establishes a model of the correlation between the chemical structure and the physiological activity. The research of 2D-QSAR focuses on two directions: the improvement of structural data and the optimization of statistical methods.
3D-QSAR is a method that introduces three-dimensional structure information of drug molecules for quantitative structure-activity relationship research. This method indirectly reflects the characteristics of non-bonded interaction between drug molecules and macromolecules. Compared with 2D-QSAR, it has a clearer physical meaning and richer information. Therefore, since the 1980s, 3D-QSAR has gradually replaced 2D-QSAR and has become one of the main methods for rational drug design based on mechanism.
Conclusion
In this paper, various computational tools for structural/ligand-based drug design (SBDD/LBDD), including virtual screening, molecular docking, and QSAR, were compared and analyzed. Advanced tools of drug design strategies, including pharmacophore modeling, fragment-based drug design, and structural similarity search, and molecular dynamics simulation, have been widely used in modern drug discovery. These computational tools have been successfully used in drug discovery and development projects alongside bioinformatics and cheminformatics tools. In recent years, AI techniques have also been developed and can be used with omics tools (Genomics/Proteomics/metabolomics…) And computational biology.