Pharma Research: How AI is transforming the Drug Discovery Process

Janifha Evangeline, Assistant Editor, India Pharma Outlook | Tuesday, 14 November 2023

 Janifha Evangeline, Assistant Editor, India Pharma Outlook
From the works of Louis Pasteur and Alexander Fleming to the present era of precision medicine and immunotherapy, the field of drug research and development (R&D) has experienced remarkable advancement in the last two centuries. In spite of these advances, there are still significant obstacles in developing effective treatment methods for complex, multifactorial illnesses. The search for dependable and informative models to successfully guide the development of drugs continues to be a pressing and ongoing concern. In recent times, there has been a remarkable surge in the popularity of transformer-based models like generative pre-trained transformers (GPT), which have shown immense potential to revolutionize various domains, including life sciences. This article delves into the potential of recent progress in artificial intelligence (AI) to transform the field of drug research and development (R&D). AI provides new opportunities for modelling biology and creating treatments by interpreting the language of biology and combining various types of data. Navigating the challenge: Drug development for complex diseases Drug development faces considerable difficulty in dealing with multifactorial diseases including cancer, neurodegenerative conditions, and inflammatory disorders. Diseases of this nature encompass intricate interactions among numerous biological mechanisms and factors. As an example, changes in the genetic makeup can impact the production of proteins which then affects how cells interact and their spatial organization. The entire web of interconnected biological relationships has an influence on various aspects, ranging from the beginning and advancement of diseases to the eventual reaction of patients towards treatment. In order to create successful treatments, it is crucial to have a thorough comprehension of how different biological systems interact with each other. Although the resources and tools for drug development have advanced significantly, effectively utilizing them has become more difficult. This task now demands diverse knowledge, skills, experiences, and the capability to integrate various data types. It is no longer a task that can be managed by individuals alone. Moreover, it should be noted that traditional approaches such as cell and animal models frequently fail to fully capture the complexities of human biology, thus diminishing their ability to accurately predict outcomes for complex diseases. Iambic Therapeutics focuses on using AI algorithms based on physics to discover molecules with unique ways of functioning. The technology it possesses can anticipate the interaction between a molecule and a disease target. Although computational, predictive models have been employed in biological research for a considerable time, existing models usually require specialized knowledge and usually focus on specific aspects of biology rather than accurately reflecting the complex nature of the disease. There is an immediate requirement for groundbreaking solutions that comprehensively address the intricate nature of biological systems and aid in the advancement of drug research and development for better and more efficient treatments.
Using machine learningto analyze human tissue data, Verge Genomics' CONVERGE platform will be employed by AstraZeneca, specifically through its rare disease division called Alexion. The aim is to find new drug targets for uncommon neurodegenerative and neuromuscular disorders. “In the future we believe that AI may help us predict what queries regulators are likely to come back with,” says Boris Braylyan, Vice President and Head of Information Management at Pfizer. “We may then be able to improve our submissions by predicting in advance what regulators are likely to ask, and coming prepared with those answers ahead of time.” From lab to the clinic: comprehensive disease modelling These applications for LLM demonstrate the potential of transformer-based models in efficiently accessing, analyzing, and making predictions based on biological data. Thanks to these advances, there is a growing possibility for the emergence of revolutionary tools. These will not only bring together our current scientific knowledge, but also combine the countless amounts of data, ranging from genomics to spatial biology, to form a thorough, interconnected, and ever-evolving disease model. Rather than researchers concentrating on separate systems or components of illnesses, this potent framework can be employed to achieve something that was previously unattainable: visualizing and simulating diverse disease processes in their entirety and interconnectedness. This all-encompassing and thorough strategy would pave the way for groundbreaking advancements in biopharmaceutical research. Instead of concentrating on specific results, screening assays in the field of drug discovery could be enhanced by comprehending the overall effect of a compound on the patient. Merck KGaA plans to increase its utilization of AI-based design and discovery abilities through partnerships with BenevolentAI and Exscientia. The objective is to improve the process of developing drugs in the fields of oncology, neurology, and immunology. Taking into account all aspects of the situation increases the likelihood of identifying drugs that are truly effective in real-life clinical settings. When studying the mechanisms of action for drugs, these models have the potential to illustrate a wide range of interconnected effects caused by a new compound. For example, if focusing on a specific pathway, these models have the ability to detect additional processes that could unintentionally be influenced or unveil surprising resistance or compensatory mechanisms – outcomes that are often overlooked using conventional methods. In the field of clinical research, they have the capability to preconceive potential negative occurrences, elucidate the reasons behind unresponsive patients to treatment, and even foresee the outcomes of modifying dosages. “We are now seeing AI used not only to further improve target discovery – but also to deliver step-change improvements in manufacturing, process efficiencies and clinical trials,” says Tara Dougal, Content Director – Pharma at Informa Markets. These observations have the potential to guide the development of clinical trials with better knowledge, customized treatment strategies, and ultimately, enhanced patient results. This comprehensive approach, which combines the knowledge gained from years of scientific research and careful experimental observations, has the potential to completely reshape the field of drug discovery and development.

© 2024 India Pharma Outlook. All Rights Reserved.