Insilico Medicine("Insilico"), a clinical-stage generative artificial intelligence (AI)-driven biotechnology company, today announced that the company has achieved a groundbreaking collaboration with Inimmune, which will utilize Chemistry42, Insilico's proprietary generative artificial intelligence (AI) technology to accelerate the discovery and development of next-generation immunotherapeutics.
Chemistry42, a multi-agent reinforcement learning system designed with medicinal chemists in mind, is designed to tackle key challenges in small molecule drug discovery, including novelty, diversity, property prediction, and multi-parametric optimization. By incorporating more than 42 advanced machine learning technologies including generative autoencoders, generative adversarial networks, and evolutionary algorithms as well as about 500 pre-trained models, Chemistry42 enables the generation and design of drug molecules with custom physicochemical properties from inception. It also supports the evaluation of multidimensional features such as pharmacological efficacy, metabolic stability, synthetic difficulty, ADME properties, and selectivity of the generated molecules.
In the initial phase of their collaboration, Inimmune leveraged the capabilities of Chemistry42 to tackle specific challenges in their drug discovery efforts. The platform's ability to generate novel template molecules and evaluate multiple key attributes, including metabolic stability, synthetic difficulty, and ADME (Absorption, Distribution, Metabolism, and Excretion) properties proved invaluable. Chemistry42 enabled the rapid generation and screening of molecules with high potential efficacy against targeted biological pathways, allowing Inimmune to efficiently identify promising lead compounds. The platform's synthetic feasibility assessment provided crucial insights, helping prioritize candidates that were both effective and synthetically accessible, further streamlining the drug development process. By integrating Chemistry42's capabilities, Inimmune significantly enhanced its research strategies and decision-making, driving greater efficiency and precision in their drug discovery efforts.
Inimmune's highly skilled chemists were able to take the software and quickly provide insights toward novel compounds with improved potency and pharmacodynamic properties. This collaboration led to significant improvements in the efficiency of Inimmune's drug discovery process. This initial success laid a strong foundation for ongoing collaboration, demonstrating the value of integrating AI-driven platforms like Chemistry42 into the drug discovery pipeline. The improvements in efficiency and the generation of high-potential hit series highlighted the transformative impact of advanced computational tools on pharmaceutical research and development.
Having completed our trial and the initial round of compound generation, we are now advancing to the synthesis and biological testing phase. We are pleased to maintain our access to Chemistry42, as it allows us to efficiently evaluate and prioritize the compounds for synthesis. We look forward to accelerating the delivery of innovative vaccines and immunotherapies for unmet medical needs through transformative AI-driven approaches that streamline the drug discovery process and improve the quality of compounds."
Ahmad Junaid, PhD., senior scientist at Inimmune
"It was a pleasure working with Inimmune," said Hugo de Almeida, PhD, Application Scientist and CADD Specialist at Chemistry42. "It's clear to me that they are very committed to incorporating state-of-the-art tools to discover new compounds and improve patients' lives. Their interest in AI and generative chemistry resulted in good communication, enabling us to identify their challenges and provide solutions to generate new ideas that will hopefully help them discover new drugs."
Building on the initial success achieved in the first round of compound generation, Inimmune researchers plan to further optimize the Hit series identified by Chemistry42. The goal is to refine these compounds through an additional 2-3 rounds of iterative optimization, focusing on enhancing their desirable properties to develop lead compounds suitable for advancement to the next stages of synthesis and testing.
During these optimization rounds, Inimmune will utilize Chemistry42's advanced algorithms to fine-tune various molecular attributes, ensuring that the lead compounds exhibit the optimal properties. By leveraging the platform's capabilities, the researchers at Inimmune also aim to overcome any challenges in current SAR campaigns and improve the overall drug-like properties of the current assets.
The significance of the collaboration lies in its potential to greatly enhance the efficiency and effectiveness of drug discovery processes. Traditional methods of drug development are often time-consuming and costly, with a high rate of failure. The use of AI and machine learning in Chemistry42 offers a promising solution to these challenges by enabling the rapid generation and evaluation of novel compounds with desirable properties. This approach not only speeds up the drug discovery process but also increases the likelihood of identifying successful therapeutic candidates.
Alan Joslyn, PhD., Inimmune CEO said, "The initial success we've seen with Chemistry42 is just the beginning. By leveraging their cutting-edge technology, we're able to refine our lead compounds further and address key challenges in our SAR campaigns, ultimately driving us closer to bringing innovative therapies to patients."
In 2016, Insilico first described the concept of using generative AI for the design of novel molecules in a peer-reviewed journal, which laid the foundation for the commercially available Pharma.AI platform. Since then, Insilico keeps integrating technical breakthroughs into Pharma.AI platform, which is currently a generative AI-powered solution spanning across biology, chemistry and clinical development. Powered by Pharma.AI, Insilico has nominated 18 preclinical candidates in its comprehensive portfolio of over 30 assets since 2021 and has received IND approval for 9 molecules.
Simple peptide modification could unlock a whole class of targets for drug discovery