Abseil is using generative AI and microbial engineering to revolutionize biologic drug discovery, increasing success rates and paving the way for personalized medicine.
The application of generative AI in antibody engineering allows for the rapid design and validation of antibodies with desired attributes, significantly enhancing the accuracy of binding interaction predictions.
The future of AI-enabled drug discovery holds promise for achieving personalized medicine, improving patient outcomes, and reducing the cost of failed drug candidates through advanced hardware, wet lab techniques, and software algorithms.
Deep dives
Advancing Biologic Drug Discovery with Generative AI
Abseil, a public company, is using generative AI to revolutionize biologic drug discovery by harnessing E.coli to produce antibodies and leveraging AI to design antibodies with specific attributes. The traditional approach of antibody discovery using mice or phage display lacks control over location, affinity, and developability of the antibodies, resulting in long timelines and low success rates in the clinic. However, Abseil's combination of generative AI, microbial engineering, and advanced assays enables them to rapidly generate and screen billions of antibodies, select those with desired attributes, and validate them in the wet lab. This approach has the potential to greatly increase success rates in drug development and pave the way for personalized medicine.
Implications for Antibody Engineering and AI in Drug Discovery
The application of generative AI in antibody engineering offers promising opportunities for improving the specificity, affinity, and development of antibodies. By using AI models to design antibodies with the desired epitope specificity, affinity, and other attributes, Abseil can accelerate the drug discovery process and significantly enhance the accuracy of predicting binding interactions. The availability of large-scale, AI-generated antibody libraries and advanced assays allows for high-throughput screening and validation of millions of AI-derived designs. This combination of AI and biology has the potential to transform the field of drug discovery, enabling the design of biologics with precise characteristics and increasing the likelihood of successful clinical outcomes.
The Future of AI-Enabled Drug Discovery
The future of AI-enabled drug discovery holds promise for achieving personalized medicine and improving patient outcomes. As AI models continue to evolve and improve, the success rates of drug development are expected to increase, enabling the targeting of smaller patient populations and reducing the cost of failed drug candidates. The integration of advanced hardware, wet lab techniques, and software algorithms will further enhance the capabilities of AI in drug discovery. Ultimately, AI-driven approaches like Abseil's are poised to revolutionize the pharmaceutical industry, allowing for the rapid design and development of highly effective biologic drugs tailored to individual patient needs.
Importance of AI Models in Drug Discovery
AI models are crucial in drug discovery as they can solve complex problems that cannot be addressed with existing assays. These models can be used to prompt specific epitopes or molecular properties, improving accuracy and reducing the number of molecules that need to be screened. The ability to benchmark and evaluate models is essential for making progress in biologic drug discovery. The use of generative AI, combined with synthetic biology, unlocks new biology and targets, particularly for challenging areas like GPCRs, ultimately benefiting patients.
Naturalness Model and Optimization in Antibody Development
The naturalness model plays an important role in antibody development, focusing on key factors that make an antibody natural and developable. By training the model on vast amounts of antibody sequences, the likelihood of a sequence being found in natural immune repertoires can be assessed. This model helps reduce immunogenicity and ensure manufacturability of antibodies, leading to successful clinical trials. The decreasing costs of DNA synthesis and sequencing contribute to the scalability of data generation, enabling more efficient and cost-effective research and design in drug discovery.
Drug development is an extremely expensive endeavor, in large part because of the 96% failure rate of clinical drug trials. However, the new frontier of medicine has arrived, and Sean McClain and Joshua Meier (respectively the Founder/CEO and Chief Artificial Intelligence (AI) Officer of Absci) are leading the charge. Absci is a public company harnessing generative AI to create more effective medicines faster and less expensively. During this episode, Sean and Joshua explain how their pooled approach to antibody production works, why they focus so heavily on data, the metrics they use to evaluate their models, and why they are certain that personalized medicine is going to be a reality in the future. There are countless possible mashups of AI and drug discovery, and Absci attempts to bring an innovative approach to biologics to positively impact human health!
