Ep. 78: “Artificial Intelligence and Immunology” Featuring Drs. Jo Viney and Ryan Peckner
May 21, 2024
auto_awesome
In this insightful discussion, Dr. Jo Viney, CEO of Seismic Therapeutic, and Dr. Ryan Peckner, Director of Machine Learning, delve into the exciting intersection of artificial intelligence and immunology. They explore the transformative role AI plays in drug discovery and protein engineering, unveiling its potential to innovate antibody development and enhance therapeutic efficacy. The duo also addresses the challenges in integrating AI into clinical trials and underscores the importance of human expertise alongside AI for reliable scientific advancements.
AI's integration into drug discovery allows for efficient design and testing of biologics to treat autoimmune diseases.
Generative AI enhances protein engineering by enabling the design of novel protein structures and optimizing antibodies and enzymes.
AI has the potential to revolutionize immunology through improved target identification, patient stratification, and preclinical testing processes.
Deep dives
The Evolution and Basics of AI and Machine Learning
AI is broadly defined as the use of computer models and algorithms to perform tasks that typically require human intelligence. This field, rooted in early theoretical work since the 1930s, has gained momentum due to significant advancements in computing power and the availability of vast amounts of data. Particularly in the last two decades, the growth of neural networks and machine learning has transformed AI capabilities, making it possible for algorithms to make predictions and generate new data. The distinction between generative and non-generative AI is crucial, with generative AI capable of creating new instances from learned representations, highlighting a pivotal shift in the practical applications of AI.
AI's Application in Drug Discovery and Biologics
AI is increasingly utilized in drug discovery, particularly in creating biologics to treat autoimmune diseases. Seismic Therapeutic focuses on leveraging machine learning to enhance protein therapeutic drug discovery, bringing together collaboration among structural biologists and machine learning experts. Strategies such as protein fitness evaluation and prediction using large datasets have emerged, making it possible to design and test new drugs more efficiently. This integration of AI not only accelerates the drug development process but also aims to ensure optimal drug-like properties, improving outcomes for patients.
The Role of Generative AI in Protein Engineering
Generative AI represents a significant advancement in protein engineering, allowing scientists to design novel protein structures by manipulating latent representations within models. By understanding and imitating the properties and functionalities of naturally occurring proteins, researchers can generate structural variants that exhibit desired functionalities. This capability also extends to optimizing antibodies and enzymes, providing a pathway to more efficient drug design. The implications of such technology foster collaboration between human expertise and AI, ultimately enhancing the success rate of drug development.
Expanding AI Utilization in Immunology and Clinical Research
AI has notable potential for broader applications in immunology, including target identification based on genomic data, patient stratification, and predicting responses to treatments. The use of AI can help analyze complex datasets from patient records, enabling researchers to identify trends and determine effective therapeutic strategies more rapidly. Moreover, efforts are underway to enhance AI capabilities in preclinical testing, potentially minimizing animal testing while maintaining high safety and efficacy standards. These advancements could revolutionize the drug discovery landscape by expediting the transition from research to clinical application.
Future of AI in Science Communication and Education
AI tools are becoming increasingly important in science communication, facilitating the writing process, detecting anomalies in research, and suggesting improvements. As AI continues to evolve, it is expected that more scientists will need to be familiar with AI applications to enhance their work. Data-driven tools can serve as valuable resources for hypothesis generation and research integration, yet they should complement rather than replace human expertise. Ultimately, the personal touch in science communication remains irreplaceable, emphasizing the need for scientists to balance AI assistance with their unique voice and perspective.
Dr. Jo Viney is the CEO, President, & Co-Founder and Dr. Ryan Peckner is the Director of Machine Learning & Computational Biology at Seismic Therapeutic. In this special episode recorded in front of a live audience at IMMUNOLOGY2024 in Chicago, they discuss artificial intelligence and its applications in drug discovery and protein engineering. They talk about its limitations, its role in clinical trials, and how scientists can leverage these tools to advance their careers.
in Chicago, they discuss artificial intelligence and its applications in drug discovery and protein engineering. They talk about its limitations, its role in clinical trials, and how scientists can leverage these tools to advance their careers.
