DeepCure Unveils AI-Driven Chemical Synthesis Platform
DeepCure, a company specializing in AI-driven therapeutic discovery, has introduced a development in chemical synthesis with its Inspired Chemistry™ platform. This technology aims to address the limitations of manual chemical synthesis in drug discovery by utilizing high-throughput, multi-step robotic synthesis. The platform is designed to facilitate the creation and testing of more complex and diverse small molecules, including those designed by AI.
In a recent demonstration, the Inspired Chemistry platform synthesized nirmatrelvir, a key component of Paxlovid, along with 56 analogs through a robot-driven workflow. This platform, now accessible to early partners, could accelerate drug discovery by enabling researchers to synthesize and evaluate a wide array of complex molecular designs more efficiently.
Traditionally, small molecule drug discovery has relied heavily on manually synthesized molecules, a process that is both costly and time-consuming. These limitations have constrained innovation and optimization in the field. Previous attempts to automate synthesis have often fallen short in terms of chemical diversity, complexity, throughput, or cost-effectiveness.
Kfir Schreiber, CEO of DeepCure, commented on the potential of the Inspired Chemistry platform:
"We are in a new phase for drug discovery with opportunities to identify small molecules for novel mechanisms and target classes. To truly harness this potential, we require a wealth of experimental data to inform the design of these new molecules. Automated synthesis is essential to quickly and cost-effectively test thousands of molecules to explore diverse chemical spaces, understand new structure-activity relationships, and optimize efficacy and safety."
The platform aims to automate the simultaneous synthesis of thousands of molecules with complex characteristics, such as multiple chiral centers. For instance, the synthesis of nirmatrelvir involves ten synthetic steps, including five purification stages, and the platform was able to produce 30 mg of the correct stereoisomer at 98% purity, along with over fifty diverse analogs at a 1 mg scale. This capability could be beneficial for drug discovery programs that require rapid generation of diverse and complex chemical matter from initial hits or to explore extensive chemical spaces around lead molecules.
Thras Karydis, CTO of DeepCure, on the broader applications of this technology:
"Our original goal was to remove the diversity constraints of commercial libraries and long iteration cycles of outsourced manual chemical synthesis from AI-based molecule design. Our platform makes chemical synthesis an enabler of drug design rather than a barrier. Automated synthesis has potential applications not only in drug discovery but also in fields ranging from OLED displays to household and agricultural products."
Topics: AI & Digital
