Structure-Based Virtual Screening

The drug development process is a major challenge for the pharmaceutical industry because it requires a lot of time and money to go through all the stages of developing new drugs. One widely used method to minimize the cost and time of the drug development process is computer-aided drug design (CADD). CADD can better focus on experiments, which can reduce the time and cost of researching new drugs. In this case, structure-based virtual screening (SBVS) is both powerful and useful, and is one of the most promising computer technologies for drug design. SBVS tries to predict the best interaction mode between two molecules to form a stable complex, and it uses a scoring function to evaluate the force of the non-covalent interaction between the ligand and the molecular target.

Principle of structure-based virtual screening

Structure-based virtual screening (SBVS), also known as target-based virtual screening (TBVS), attempts to predict the optimal interaction between ligands to form complexes against molecular targets. As a result, the ligands are ranked according to their affinity to the target, and the most promising compounds are shown at the top of the list. The SBVS method requires knowing the 3D structure of the target protein so that the interaction between the target and each compound can be predicted by the computer. In this strategy, compounds are selected from a database and classified according to their affinity for the receptor site. In SBVS technology, molecular docking is worth noting because of its low computational cost and good results.

Overall solutions

First we will design one or more goals. Usually, these targets are closely related to specific diseases, and inhibiting or changing these targets will help treat the disease. Design and synthesize a compound library containing various small molecule drug fragments. The ideal small molecule fragment should have a certain affinity to the active site of the target, such as a target protein or a targeted enzyme. Use pre-designed targeting targets to screen small fragments of synthetic fragments. According to the screening results, molecular fragments with better activity can be obtained, and lead compounds can be obtained by appropriately combining the fragments. Further modify and further optimize the chemical structure of the obtained lead compound to obtain candidate drugs for clinical research.

AI-based approaches

The method based on AI learning is considered to be an important scoring function. AI-based learning methods have attracted attention because of their reliable predictions. Many researchers have used AI to improve the SBVS algorithm. These techniques have been used in quantitative structure-activity relationship (QSAR) analysis to predict various physical chemistry (for example, molecular hydrophobicity and stereochemistry), biology (for example, activity and selectivity) and drugs (for example, absorption and Metabolism) characteristics. Among these types of scoring functions, modern QSAR analysis can be used to derive statistical models to calculate protein-ligand binding scores.AI-based algorithms include:

• Artificial neural networks
• Support Vector Machines
• Bayesian technique
• Decision tree
• k nearest neighbor (kNN)
• Kohonen's SOM and backpropagation ANN
• Integrated approach using machine learning