XtalPi is dedicated to developing its Intelligent Digital Drug Discovery and Development (ID4) platform for pharmaceutical research. Combining its expertise in quantum mechanics, artificial intelligence, and cloud computing, XtalPi improved upon existing pharmaceutical research methods to offer innovative R&D technologies that can accelerate drug discovery and increase pipeline success rate, including general force field development for drug-like molecules, high-accuracy binding affinity prediction, and AI-based molecule generation.. XtalPi provides a variety of drug design solutions such as hit identification, hit-to-lead/lead generation, and lead optimization. Our methods are capable of generating millions of new candidate compounds for desired targets, accurately predict their properties like the biological activity, target selectivity, and druggability. XtalPi helps our clients to accelerate the small-molecule drug discovery and development, reduce R&D costs, and improve the overall success rate of projects. XtalPi has developed the cloud-based platform to support our ID4 platform, where over a million cores of computing power ensures maximum security, scalability, flexibility, and efficiency. Based on that, we provide a variety of automated, efficient, and intelligent drug discovery services to world-wide pharmaceutical companies.
For a specific target, the first step in the drug discovery process is the identification of hit compounds. Small molecules with novel core structures and reasonable biological activity can be obtained through methods such as high-throughput screening, virtual screening, and de novo design. The promising hit series then move into the next lead-generation phase.
The next key step in the drug discovery process is lead generation, also known as hit-to-lead optimization. Starting from the hit compounds emerged from the earlier stage, a combination of scaffold hopping, R-group substitution, structural simplification, and molecular hybridization can be used to modify the structures of hits to achieve better potency, selectivity, and drug-like properties simultaneously. The main focus of this step is to yield at least one lead compound for the subsequent optimization.
The last step before arriving at a clinical drug candidate is the optimization of lead compounds, which is also an extremely complicated process that requires multi-parameter optimization within a patentable chemical space. At this stage, the structures of the drug candidates are modified to fully enhance their selectivity, physicochemical and ADMET properties, while maintaining their biological activity. Eventually, a safe and effective preclinical candidate compound (PCC) which is ideal for clinical tests will be obtained. The traditional lead optimization process usually costs a significant amount of time and resources, no matter whether it is related to the development of me-too, me-better or first-in-class drugs. Therefore, it is critical to bring in more efficient and innovative methods to speed up this process.