Revolutionizing Medicine: A.I. in Drug Development

⏱️ 17 min read       

Our world continues to constantly dive into new cutting-edge innovations shaping our world. One of those innovations is happening now in various high-tech labs where scientists are harnessing the power of artificial intelligence to revolutionize drug development. Imagine a future where life-saving medicines are designed faster and more effectively than ever before. 

This transformation is not only just getting started today, but happening at a pace that will conduct a complete makeover for the drug development world. A.I. is reshaping the future of healthcare. 

Welcome to That's Life, I Swear.  This podcast is about life's happenings in this world that conjure up such words as intriguing, frightening, life-changing, inspiring, and more. I'm Rick Barron your host. 

That said, here's the rest of this story  

At the heart of Monrovia, [L.A. area] California, a cutting-edge facility hums with activity. This is no ordinary workspace; it's Terray Therapeutics data powerhouse, where the future of medicine is being manufactured.

Picture a vast space, nearly as large as a football field, filled with an intricate display of machines and humans. Here, the traditional image of test tubes and Bunsen burners gives way to a high-tech arena where robots and computers reign supreme.

The true magic, however, unfolds on a scale invisible to the naked eye. Imagine microscopic "muffin tins" etched into silicon, each depression a stage for a molecular ballet. In these tiny wells, chemicals and proteins meet, their interactions carefully recorded. The sheer volume of data generated daily is staggering – equivalent to over 12,000 full-length films.

This isn't just a laboratory; it's the vanguard of a revolution in pharmaceutical research. Terray Therapeutics is part of a growing movement of innovative firms harnessing the power of artificial intelligence to revolutionize drug discovery.

Gone are the days of purely intuitive, trial-and-error approaches. These companies are ushering in an era where massive data sets feed sophisticated AI systems, dramatically accelerating the drug development process. As Jacob Berlin, Terray's co-founder and CEO, puts it, "Feed the AI the right data, and its capabilities become truly remarkable."

Jacob Berlin and his brother, Eli, co-founders of Terray Therapeutics. Courtesy of The New York Times

In this new paradigm, the art of drug discovery is being transformed into a precise, data-driven science, promising faster breakthroughs and more effective treatments for the future.

Why does this matter?

In the rapidly evolving landscape of artificial intelligence, a groundbreaking application is emerging that could revolutionize the pharmaceutical industry. While many early adopters of generative AI have focused on streamlining office tasks and enhancing customer service, the realm of drug discovery stands poised for a transformative leap forward.

Industry experts, including renowned consulting firm McKinsey & Company, are praising AI as a pivotal advancement in pharmaceutical research, potentially ushering in a new era of medical breakthroughs. This technology's impact on drug development could be so profound that it's being looked at as a once-in-a-century opportunity.

At the heart of this AI-driven approach to drug discovery lies a foundation of highly specialized data. Unlike the broad-based training of popular AI models such as ChatGPT or DALL-E, which draw from vast internet resources, pharmaceutical AI systems are built upon intricate scientific information. This includes detailed molecular structures, protein configurations, and complex biochemical interactions.

The AI's role in this process is something along the lines to a sophisticated matchmaking system, analyzing patterns within this specialized data to identify promising drug candidates. It's as if the AI is searching for the perfect chemical key to unlock key protein targets, potentially leading to new treatments for various diseases.

One significant advantage of this tailored approach is the reduced risk of erroneous outputs or "hallucinations" that can plague more generalized AI systems. The precision and relevance of the scientific data used in pharmaceutical AI contribute to more reliable and actionable results.

Hallucinations defined:
In the field of artificial intelligence (AI), a hallucination or artificial hallucination (also called bullshitting, confabulation or delusion) is a response generated by AI which contains false or misleading information presented as fact. This term draws a loose analogy with human psychology, where hallucination typically involves false percepts. However, there is a key difference: AI hallucination is associated with unjustified responses or beliefs rather than perceptual experiences. [Wikipedia]

However, it's crucial to note that AI-suggested drug candidates are just the beginning of a long and rigorous journey. Any potential new drug must undergo extensive laboratory testing and clinical trials before it can be considered for patient use. This ensures that the power of AI is harnessed responsibly within the established framework of pharmaceutical safety and efficacy standards.

Data is power

In the cutting-edge realm of pharmaceutical development, firms like Terray are revolutionizing the landscape with expansive, high-tech laboratories. These facilities serve as data powerhouses, generating vast troves of information to fuel artificial intelligence systems. This union between data and AI enables rapid experimentation, pattern recognition, and predictive modeling of potential therapeutic compounds.

Technicians working at the Terray office. Courtesy of The New York Times

The process begins with generative AI algorithms crafting digital blueprints for drug molecules. These virtual designs are then brought to life in state-of-the-art automated labs, where they're synthesized and evaluated for their interactions with target proteins. Each result, whether promising or disappointing, feeds back into the AI system, refining its understanding and improving future designs. This iterative cycle dramatically accelerates the traditionally arduous drug discovery process.

While AI-designed drugs have begun entering clinical trials, it's important to note that we're still in the early stages of this technological revolution. As David Baker, a renowned biochemist and director of the Institute for Protein Design at the University of Washington, aptly puts it: "Generative AI is indeed transforming the field, but we must remember that drug development remains an intricate and deeply human endeavor."

This balanced perspective underscores both the immense potential and the enduring complexities of merging artificial intelligence with the nuanced world of pharmaceutical research.

The pharmaceutical industry has long struggled with a daunting reality: drug development is a high-stakes gamble, along with astronomical costs and protracted timelines. 

·       While estimates vary, the average price tag for steering a single drug from conception to market approval hovers around $1 billion 

·       This laborious journey typically spans 10 to 15 years 

·       High failure rates of nearly 90% for candidate drugs entering human clinical trials, often due to inefficacy or unexpected side effects.

Enter the new wave of AI-powered drug developers, armed with cutting-edge technology and a bold vision to rewrite these discouraging statistics. Their mission: to dramatically improve success rates while simultaneously slashing both time and financial investment.

These AI trailblazers have found a lifeline in big pharma companies, which have historically played the dual role of partner and financier to smaller research ventures. Today's AI-driven firms are laser-focused on revolutionizing the preclinical stages of development, a phase that has traditionally consumed four to seven years. While some may venture into clinical trials independently, this costly and complex stage is typically where pharmaceutical giants step in, leveraging their vast resources to navigate the lengthy human trials process, which can stretch an additional seven years.

This symbiotic relationship between AI innovators and established pharma powerhouses represents a new frontier in drug discovery, one that holds the promise of bringing life-saving treatments to patients faster and more efficiently than ever before.

In the rapidly evolving landscape of pharmaceutical research, established drug companies are carefully aligning themselves with innovative partners, viewing this approach as a cost-effective means to harness cutting-edge advancements.

This strategy for pharmaceutical giants can best be described as, calling for an Uber rather than purchasing a vehicle outright. They gain access to innovation without the hefty price tag of in-house development.

This symbiotic relationship is structured around milestone-based payments, with pharmaceutical giants rewarding their research partners for achieving key benchmarks in drug candidate development. These payments can accumulate to hundreds of millions of dollars over time. Should a drug ultimately receive approval and achieve commercial success, a continuous stream of royalty income further sweetens the deal for the innovative partners.

Who are the current leaders?

At the forefront of this revolution are companies like:

·       Terray

·       Recursion Pharmaceuticals

·       Schrödinger

·       Isomorphic Labs    

Each are striving for groundbreaking advancements. However, their approaches diverge into two distinct routes: those investing in expansive laboratory facilities and those opting for a more virtual model.

This dichotomy in strategy reflects the diverse ways in which these pioneering firms are leveraging technology and resources to reshape the future of drug discovery and development.

In the realm of AI-driven drug discovery, Isomorphic Labs, a spinoff from Google DeepMind, is charting a unique course. Their philosophy hinges on the premise that superior AI can extract more value from less data, placing their bets squarely on software excellence.

The year 2021 marked a watershed moment when Google DeepMind unveiled groundbreaking software capable of accurately predicting protein structures from amino acid sequences. This advancement was a game-changer, offering deeper insights into biological processes and accelerating drug discovery efforts by making clear the intricate relationship between protein shape and function.

Building on this success, Google DeepMind and Isomorphic recently introduced AlphaFold 3, an AI model that takes protein prediction to new heights. This cutting-edge system can forecast interactions between molecules and proteins, representing another leap forward in the intricate dance of drug design.

Max Jaderberg, Isomorphic's chief AI officer, underscores their strategy: "We're doubling down on computational approaches. We believe there's an untapped wellspring of potential waiting to be unleashed."

While Isomorphic forges ahead with its software-centric approach, companies like Terray represent a different facet of the AI drug development landscape. These startups are the product of years of scientific research, now turbocharged by recent breakthroughs in artificial intelligence.

This diverse ecosystem of approaches highlights the multifaceted nature of AI's role in revolutionizing drug discovery and development.

At the helm of Terray stands Dr. Berlin, a visionary leader whose journey through the realms of nanotechnology and chemistry began with a Ph.D. from Caltech. His career-long pursuit of scientific breakthroughs has culminated in Terray, a company born from the seeds of an academic project initiated over a decade ago at the City of Hope cancer center near Los Angeles, where Dr. Berlin led a pioneering research group.

Terray has set its sights on a specific niche within the vast pharmaceutical landscape: small-molecule drugs. These are the everyday heroes of medicine, the pills we swallow without a second thought, like aspirin or statins. The appeal of these drugs lies in their convenience for patients and their cost-effectiveness in production.

Statins defined:

Statins (or HMG-CoA reductase inhibitors) are a class of medications that reduce illness and mortality in people who are at high risk of cardiovascular disease. They are the most commonly prescribed cholesterol-lowering drugs. [Wikipedia]

The contrast between Terray's cutting-edge facilities and the basic tools of yesteryear's academic research is stark. Gone are the days of data languishing in Excel spreadsheets and automation existing only as a distant dream.

Kathleen Elison, a co-founder and senior scientist at Terray, offers a glimpse into this transformation. "I was the robot," she said, recalling the labor-intensive methods of the past. Her words underscore the quantum leap from manual processes to today's high-tech, AI-driven drug discovery paradigm.

Kathleen Elison, a co-founder. Courtesy of The New York Times

This evolution from academic roots to sleek, automated labs embody the rapid acceleration of drug development technology, promising a future where life-saving treatments may emerge faster and more efficiently than ever before.

When Terray emerged in 2018, the technological landscape was ripe for revolution. The company seized upon advancements in micro-scale chip manufacturing, leveraging external expertise to bring their custom designs to life. Their labs, a testament to modern ingenuity, are filled with customized automated equipment, made possible by leaps in 3-D printing technology.

From its inception, Terray's visionaries recognized artificial intelligence as the key to unlocking the potential of their vast data repositories. However, the true power of generative AI in drug development only became apparent as the field evolved - predating even the ChatGPT phenomenon that captivated the world in 2022.

In 2020, Narbe Mardirossian, formerly a senior scientist at Amgen, joined Terray as Chief Technology Officer. His decision was influenced significantly by the company's wealth of laboratory-generated data - a goldmine for AI applications. Under Mardirossian's guidance, Terray has significantly expanded its data science and AI capabilities, developing a groundbreaking AI model that seamlessly translates between chemical data and mathematical representations. In a move towards scientific openness, the company has made an open-source version of this model available to the wider research community.

This journey from concept to cutting-edge AI implementation illustrates Terray's commitment to pushing the boundaries of drug discovery. By harnessing the power of data, customized technology, and artificial intelligence, Terray is paving the way for a new era in pharmaceutical research - one where the next breakthrough could be just an algorithm away.

Terray's innovative approach has not gone unnoticed in the pharmaceutical world. The company has forged key partnerships with industry giants Bristol Myers Squibb and Calico Life Sciences, an Alphabet subsidiary focused on age-related diseases. While the specifics of these collaborations remain under wraps, they signal strong industry confidence in Terray's potential.

The company's ambitious expansion plans, however, will require capital beyond its initial $80 million dollars in venture funding. Eli Berlin, younger brother of Dr. Berlin and co-founder, left a lucrative career in private equity to serve as Terray's chief financial and operating officer. His decision was driven by a conviction that the company's groundbreaking technology could unlock unprecedented business opportunities.

Terray's current focus lies in developing novel treatments for inflammatory conditions such as: 

·       lupus 

·       psoriasis 

·       rheumatoid arthritis 

Dr. Berlin projects that the company will have drug candidates in clinical trials by early 2026, a testament to their accelerated development process.

While Terray and its contemporaries are indeed revolutionizing drug discovery, Dr. Berlin maintains a pragmatic outlook. "The true measure of our success, and indeed that of the entire field, will be evident in a decade," he posits. "If we can look back and see significantly higher clinical success rates and tangibly improved human health outcomes, then we'll know we've made a real difference."

This statement captures the long-term vision driving Terray's innovations - a future where drug development is not just faster, but fundamentally more effective in addressing human health challenges. 

What can we learn from this story? What's the takeaway?

Widespread impact: The effects of AI in drug development are expected to dramatically reshape the entire healthcare industry.

This story underscores the profound impact that AI is having and will continue to have on our world, particularly in areas that directly affect human health and well-being. 

It suggests that staying informed about and adapting to these technological advances will be crucial for individuals and organizations alike.

Well, there you go, my friends; that's life, I swear

For further information regarding the material covered in this episode, I invite you to visit my website, which you can find on Apple Podcasts/iTunes for show notes and the episode transcript.

As always, I thank you for the privilege of you listening and your interest. 

Be sure to subscribe here or wherever you get your podcast so you don't miss an episode. See you soon.