Target-based drug discovery has successfully produced targeted drugs. However, it is less effective for diseases with complex pathogenic mechanisms (such as cancer, inflammation, diabetes, and central nervous system diseases). By modulating multiple targets to achieve the desired physiological response, multivariate pharmacology is emerging as a new paradigm for the treatment of complex diseases. Drug molecules that can simultaneously modulate multiple targets are a simple method for network control. As the risk of drug-drug interactions decreases, the reduction of pharmacokinetics and drug interactions requires safety profile testing. Multi-target drugs can also circumvent drug resistance caused by single target mutations or expression changes, because it is rare that multiple targets in different pathways or cascade pathways have simultaneous mutations.
Figure 1. Multiple Targeting Design.( Weilin, Zhang,et al. 2017.)
One of the key challenges in multi-target drug design is to determine feasible target combinations. In addition to systematic high-throughput screening (HTS), MedAI can also use network analysis methods. Understanding the overall topology and dynamics of the disease network can provide valuable information that can be used to identify potential therapeutic interventions and operations. In addition, we use a three-node enzyme network as a model system to study the impact of multi-target interventions. We found that drug combinations only have synergistic or antagonistic effects in certain network topologies. The dynamic behavior of the network is also very useful. For a network with realistic dynamics, the phenotypic response may be related to the network status. Network models can be used to identify potential interventions through multiple drug-target interactions that drive the network from a diseased state to a healthy state.
Pharmacology-based and docking-based virtual screening are two calculation methods commonly used in single target drug discovery. MedAI has expanded many related tools to find ligands with desired biological characteristics. A pharmacophore model with several key features can be established for each target based on the chemical structure of the known ligand or the 3D structure of the binding site. The multiple conformations of the virtual ligand are mapped to the pharmacophore model, and the suitability is evaluated. For docking-based methods, the ligand is placed into the binding site and then evaluated using different scoring functions.
MedAI provides two methods:
Figure 2. Common pharmacophore-based multi-target drug design. ( Weilin, Zhang,et al. 2017.)
AI-based multi-target drug design is a compound optimization tool for multi-objective optimization and automatic iteration for medicinal chemists and drug designers. It can start from the structural formula of a compound (seed) and generate batches of compounds through structure conversion rules. Multi-objective property evaluation will automatically pass Pareto-optimal and select N compounds with the best comprehensive properties, and then the system will automatically use these N compounds as seeds, and iteratively generate batches that meet expectations. Compound. Combined with other modules, it can evaluate and screen out compounds that meet ADMET properties, bioavailability and other objectives. It can reduce the probability that the ab initio design algorithm often encounters problematic compounds.
|Project name||Multiple Targeting Design|
|Product delivery mode||The simulation results provide you with the raw data and analysis results of molecular dynamics.|
In the past decade, the field of computational multi-objective drug design has developed rapidly. MedAI can provide you with a variety of calculation methods. In most cases, multiple target combinations can be used to control the disease network. These multiple target combinations will enable researchers to choose targets that are easily regulated by small molecules while achieving the same level of network control. If you have any needs in this regard, please feel free to contact us.
Please submit a detailed description of your project. We will provide you with a customized project plan to meet your research requests. You can also send emails directly to email@example.com for inquiries.
Tech Trends in Artificial Intelligence What's up, what's down, what lies ahead, only found in MedAI!