Leveraging AI for Drug Discovery
Accelerating drug discovery for a faster, smarter future
The future of life sciences is hurtling towards a digital revolution where the emphasis on immediacy has become crucial for the pharmaceutical, biopharmaceutical, and medical technology value chain.
“New technologies are needed following the realization of the rigidity of the supply chain during the pandemic along with the need to onshore or reshore manufacturing,” asserted Rajiv Anand, founder and CEO of Quartic.ai. “The other catalyst will be the IRA, giving the industry two headwinds that can only be dealt with through technology.”
Quartic.ai has developed a platform that applies machine learning algorithms to identify the potential efficacy of drug candidates. The system analyzes data from various sources, such as gene expression, protein interactions, and chemical structure. The program also provides insights into the mechanism of action of potential drugs, which can help researchers design better treatments.
Providing prompt and efficient solutions to patients is the sector's current main challenge and objective. In this context, innovation has emerged as a leading contributor to speeding up processes and reducing costs, thereby broadening the range of potential solutions for preventing and addressing diseases.
“Precision Oncology is about predicting the likelihood of the success of these potential therapies.”
Raymond Vennare, CEO, Predictive Oncology
Although the industry has historically shown a slow adaptation to changes, the Covid-19 pandemic has been the catalyst for accelerating the sector toward greater automation of processes. Human skills and technology are merging their functions to provide personalized solutions for patients in need. The nuances of data, cloud technology, Software-as-a-Service (SaaS) systems, artificial intelligence (AI), and machine learning (ML) algorithms are empowering life sciences companies' success and better health outcomes.
“There is a practical reason that the application of AI in drug discovery and development has been slow to be adopted,” explained Raymond Vennare, CEO of Predictive Oncology, a platform that uses AI to develop personalized cancer treatments by analyzing genomic and clinical data to identify the most effective treatments. “Until there was a critical mass of evidence, at least as applied to drug discovery, scientists were cautious. Regulatory compliance has always been an issue. Until the FDA was able to understand how best to evaluate and regulate as a medical device and decision support tool, it would have been virtually impossible to develop drugs using that technology.”
“Most AI is open source and based on big data algorithms, and that is all machine learning. The issue is that when you are in PD, you are developing a drug, and do not have much historic data, so how do you use AI?”
Rajiv Anand, Founder and CEO, Quartic.ai
Powering actionable data
New technologies have transformed many industries, and drug discovery is no exception. Biopharmaceutical companies are leveraging data to enhance site selection for clinical trials, as well as supply chain oversight, with impressive results. According to Deloitte's last Global Life Sciences Outlook, the total investment in digital therapeutics in 2023 has topped US$600 million. Despite the challenges in terms of financing, geopolitics and regulation, companies like Quartic.ai, Predictive Oncology, WhizAI, and Apprentice.io are leading the way in leveraging AI for drug discovery and development.
One of WhizAI's focuses has been on making information and analytics easily accessible and consumable for people in various roles, such as marketing, sales and finance. WhizAI has developed a generative AI machine learning model called Narratives that can not only create charts but also describe their content. “This simplifies the process of understanding and using the data, allowing users to quickly tap into valuable insights," posited Rohit Vashisht, co-founder and CEO of WhizAI.
“The potential of AI is vast, and it is just the beginning.”
Rohit Vashisht, Co-Founder and CEO, WhizAI
AI offers several benefits over traditional drug discovery methods. “AI has been embraced, as is seen with a lot of the deals being done now. AI is so broad; it can be leveraged throughout the drug discovery process to assess genomic data and find new targets, to find novel treatments and improve the properties of compounds using computational chemistry, to mine the literature to generate new hypotheses, as well as supporting systems and operations to improve workflow and efficiency,” highlighted Emer Leahy, president and CEO of PsychoGenics.
However, despite these benefits, there are still challenges to sourcing AI in drug discovery. “There is hype around AI that led to several claims that are not yet validated. Our phenotypic approach has been validated: We have clinical data on multiple compounds that confirm the pre-clinical predictions we made,” expanded Leahy.
PsychoGenics has taken a different line to drug discovery, using a phenotypic approach that focuses on the behavior of cells, tissues and organisms in response to drugs.
Developing and deploying AI-powered drug discovery tools requires careful consideration of both regulatory and ethical factors, and also needs the availability of high-quality data for successful implementation. Nonetheless, these obstacles are exciting opportunities for researchers to develop innovative solutions and unlock the highest potential of AI in drug discovery.
Article header image by Brittany Hosea-Small, courtesy of QB3 Bakar Labs