Zhu Weiliang research group and collaborators of Shanghai Institute of medicine, Chinese Academy of sciences develop a new method and network application platform D3 pockets to analyze the dynamic characteristics of drug combination pockets

Drug molecules generally play a role by binding to the target protein There are many cavities in proteins at the same time, among which the cavities that can combine with small molecules and play a role in regulating protein function are called drug-resistant pockets Different from crystalline state, the protein structure in organism always moves and changes constantly Therefore, the location and dynamic properties of drug binding pocket on protein are very important for the structure-based drug design and lead compound optimization It is reported that currently there are classification models based on scoring function or machine learning, which can be used for the prediction of drug-resistant bags However, most of these methods only focus on the static structure of the pocket, and can not fully describe the changes of the pocket in the process of protein movement In order to solve this problem, Zhu Weiliang group of Shanghai Institute of medicine, Chinese Academy of Sciences has developed a new method to study the dynamic characteristics of the pocket based on the molecular dynamics simulation trace file or conformation set, and developed the corresponding application platform D3 pockets, which is used to study the dynamic characteristics of the potential drug binding pocket on protein quality The test results of this method on HIV-1 protease, BACE1, l-abp, GPx4 and GR systems show that D3 pockets can help researchers in the field of drug design and pharmaceutical chemistry to study and compare the stability (Figure 1), continuity (Figure 2) and correlation (Figure 3) of multiple potential binding pockets on the target protein, and then select the optimal pocket to carry out active compound discovery and structural optimization Work For ease of use, the team also established a web server (http://www.d3pharma.com/d3pocket/index.php) that can submit tasks and download forecast results over the Internet The above research results were published in J chem Inf model, entitled "D3 pockets: a method and web server for systematic analysis of protein pocket dynamics" This research work is supported by national key R & D projects and other funds Xu Zhijian and Jiye Shi are co authors of this paper Figure 1 Stability analysis of HIV-1 and BACE1 protein pockets (source: J chem Inf model) Figure 2 Continuity analysis of HIV-1 and l-abp protein pockets (source: J chem Inf model) Figure 3 Correlation analysis of GPx4 and GR protein pockets (source: J chem Inf model)