How AI is accelerating drug discovery - Nature's Building Blocks | BBC StoryWorks

BBC StoryWorks
31 Aug 202304:15

Summary

TLDRAlex Zhavoronkov, founder and CEO of Insilico Medicine, discusses the transformative potential of AI in drug discovery. He highlights how AI, driven by mathematics and neuroscience, accelerates drug development, reduces risk, and improves clinical trial success. Insilico has successfully demonstrated AI's effectiveness, moving from disease hypotheses to clinical trials with reduced costs and time. Zhavoronkov envisions a future where AI democratizes biotech, enabling new entrants to innovate, and predicts personalized drug discovery tailored to individuals within 20 years, marking a significant shift in the pharmaceutical industry.

Takeaways

  • 🌍 The speaker, Alex Zhavoronkov, founder and CEO of Insilico Medicine, spends 100% of his time traveling and doesn't own a home.
  • 🧠 Insilico Medicine specializes in AI-powered drug discovery, aiming to accelerate drug development processes.
  • 📈 Around 2013-2014, deep neural networks became powerful enough to outperform humans in tasks like image and text recognition.
  • 🔮 AI is often perceived as something magical, but it is fundamentally based on mathematics and neuroscience.
  • 💡 For Alex, AI represents hope and a tool for tackling challenging biological and chemical problems in drug discovery.
  • 💊 The pharmaceutical industry is extremely risky, with most efforts failing, and requires resilience to succeed.
  • 🏆 Insilico Medicine has demonstrated success in using AI to go from disease hypothesis to clinical trials at a fraction of the cost and time.
  • ⏳ AI-driven systems, developed between 2019-2020, are now capable of generating novel molecular structures with desired properties.
  • 🚀 AI will help democratize drug discovery, allowing new entrants from emerging countries to innovate and participate in biotech.
  • 🔬 In the future, AI may lead to personalized drug discovery, where treatments are tailored to individual patients.

Q & A

  • Who is Alex Zhavoronkov and what is his role?

    -Alex Zhavoronkov is the founder and CEO of Insilico Medicine, a company specializing in AI-powered drug discovery.

  • What is the primary focus of Insilico Medicine?

    -Insilico Medicine focuses on developing AI solutions to accelerate the discovery and development of new drugs.

  • How did deep neural networks evolve around 2013-2014, according to Alex?

    -Deep neural networks became powerful enough to outperform humans in tasks such as image and text recognition, marking a major advancement in AI.

  • How does Alex describe the public perception of AI?

    -Alex states that the public often views AI as something magical, with the potential to replace human intelligence in daily tasks.

  • What does AI represent to Alex Zhavoronkov?

    -To Alex, AI represents hope. It is a tool that enables faster drug discovery, tackles challenging biological and chemical problems, and reduces risks in the pharmaceutical industry.

  • What are the main challenges in the pharmaceutical industry?

    -The pharmaceutical industry is extremely risky, requiring decades to develop new drugs. Failure is more common than success, making perseverance crucial.

  • What has Insilico Medicine achieved in AI-powered drug discovery?

    -Insilico Medicine has demonstrated success in going from disease hypothesis to clinical trials, validating their AI solutions and significantly reducing time and costs in drug discovery.

  • How did Insilico Medicine use AI to innovate drug discovery?

    -Insilico Medicine used AI to uncover novel biological targets, generate new molecular structures, and validate therapeutic programs, speeding up the process and lowering costs.

  • What is the significance of Insilico Medicine's Pharma.AI update on the 14th?

    -The Pharma.AI update showcases advancements in AI-powered drug discovery, emphasizing the potential for faster, more democratized biotech innovation.

  • What is Alex's vision for AI in the future of drug discovery?

    -Alex envisions AI systems enabling personalized drug discovery, where treatments will be tailored to individual needs, revolutionizing healthcare in the next 20 years.

Outlines

00:00

🚀 Life on the Move: A CEO’s Commitment to Innovation

Alex Zhavoronkov, the founder and CEO of Insilico Medicine, introduces himself as someone who dedicates all his time to work, with no fixed home or family. He only sleeps a few hours a day and devotes his life to advancing AI-powered drug discovery. His focus is on developing artificial intelligence solutions that accelerate the process of discovering and developing new drugs, reflecting his deep commitment to innovation and the future of medicine.

🤖 The Rise of AI in Drug Discovery

Zhavoronkov describes the transformative power of deep neural networks around 2013-2014, which began outperforming humans in tasks like image and text recognition. He explains that AI, often perceived as magical, is actually rooted in mathematics and neuroscience. For him and his team, AI symbolizes hope and is a crucial tool for tackling challenging biological and chemical problems with greater confidence, reducing risk, and increasing the probability of success in clinical trials.

💊 The Risks and Rewards of the Pharmaceutical Industry

Highlighting the inherent risks in the pharmaceutical industry, Zhavoronkov notes that developing a new drug can take decades, with a higher likelihood of failure than success. He compares the process to being a 'mixed martial arts fighter' in the pharma world, where resilience is key. Despite these challenges, Insilico Medicine has successfully demonstrated the ability to progress from disease hypothesis to clinical trials, validating their AI through their own therapeutic programs and achieving significant milestones with reduced costs and time.

🔬 Achieving End-to-End Drug Discovery with AI

Zhavoronkov outlines Insilico Medicine's achievements in drug discovery, where they have utilized their AI to go from identifying novel targets to generating novel chemistry with desired properties. This comprehensive approach has allowed them to complete the entire drug discovery process at a fraction of the usual cost and time. He also notes that many companies began adopting similar end-to-end systems around 2019-2020, but Insilico Medicine was ahead in applying these technologies for significant breakthroughs.

📅 Upcoming Pharma.AI Update: Transforming the Future

On September 14th, Insilico Medicine will release an update for Pharma.AI, their platform for AI-driven drug discovery. Zhavoronkov envisions a future where AI not only de-risks and democratizes drug discovery but also allows new entrants, including countries with no prior biotech experience, to join the innovation community. He believes that in 20 years, AI will enable personalized drug discovery, tailoring medications to individual needs, marking a revolutionary change in healthcare.

Mindmap

Keywords

💡AI-powered drug discovery

AI-powered drug discovery refers to the use of artificial intelligence to accelerate and optimize the process of finding new drugs. In the video, Alex Zhavoronkov explains that AI is being used to tackle complex biological and chemical problems with higher confidence, reducing the risks and time associated with drug discovery. This technology allows for faster identification of disease targets and the development of novel therapeutic solutions.

💡Deep neural networks

Deep neural networks are a class of machine learning algorithms modeled after the human brain’s neural connections. They have become increasingly powerful since 2013, surpassing humans in tasks like image and text recognition. Zhavoronkov highlights their role in advancing AI capabilities, which enables breakthroughs in fields like AI-driven drug discovery, crucial to the work done at Insilico Medicine.

💡Mathematics and neuroscience

Mathematics and neuroscience form the foundation of artificial intelligence, as mentioned by Zhavoronkov. AI is often perceived as magical, but he clarifies that it is rooted in mathematical algorithms and neuroscience principles. This insight demystifies AI and connects it directly to its application in understanding biological systems, particularly in the context of drug development.

💡Insilico Medicine

Insilico Medicine is the company founded and led by Alex Zhavoronkov, specializing in AI-powered drug discovery. The company focuses on developing AI solutions to accelerate the drug discovery process, from disease hypothesis to clinical trials. The video centers on the company's mission to reduce the cost and time of drug development using AI.

💡Drug discovery

Drug discovery is the process of identifying new medications based on biological targets. Zhavoronkov discusses how traditionally it is risky, costly, and time-consuming, taking decades to bring a drug to market. AI, however, is helping to de-risk this process by identifying disease targets and generating novel chemical compounds faster and more accurately.

💡Clinical trials

Clinical trials are the final stage of drug development, where new therapies are tested on humans to determine their safety and effectiveness. Zhavoronkov emphasizes how AI can increase the probability of success in clinical trials by ensuring that drugs developed through AI-based systems have already been optimized for efficacy and safety, thus reducing the risk of failure.

💡Novel targets

Novel targets refer to previously unexplored biological pathways or molecules that can be targeted for therapeutic intervention. In the context of AI-powered drug discovery, these targets are identified using advanced AI algorithms. Insilico Medicine uses AI to uncover and validate such targets, which are key to developing new treatments.

💡Pharmaceutical industry

The pharmaceutical industry is the sector responsible for the development, production, and marketing of medications. Zhavoronkov explains that this industry is highly risky, with most drug discovery efforts failing. However, AI is being utilized to de-risk these efforts by improving the identification of targets and optimizing drug candidates faster and more efficiently.

💡Personalized drug discovery

Personalized drug discovery is the future goal of tailoring medications to individual patients based on their genetic, environmental, and lifestyle factors. Zhavoronkov believes that AI will enable this within the next 20 years, allowing for more precise and effective treatments that are customized to each person’s unique biological makeup.

💡Pharma.AI

Pharma.AI is a platform developed by Insilico Medicine, designed to streamline the drug discovery process using AI. The video mentions the upcoming release of an update to this system, which will further enhance the ability of AI to generate molecular structures with desired properties. Pharma.AI represents the intersection of AI and biotechnology, pushing the boundaries of what’s possible in drug development.

Highlights

Alex Zhavoronkov is the founder and CEO of Insilico Medicine, specializing in AI-powered drug discovery.

Alex travels 100% of the time, doesn't own an apartment, and considers Insilico his life.

Deep neural networks became powerful around 2013-2014, outperforming humans in tasks like image and text recognition.

AI is often perceived as magical and may replace human intelligence in daily tasks, which is not far from the truth.

AI represents hope and is a tool to accelerate drug discovery, tackle challenging biological and chemical problems, and increase clinical trial success rates.

The pharmaceutical industry is extremely risky; it can take decades to bring a new drug to market, with more failures than successes.

At Insilico, they have demonstrated the ability to go from disease hypothesis to clinical trials with high probability of success.

Insilico Medicine has validated its AI using its own therapeutic programs, generating novel chemistry with desired properties.

Insilico’s end-to-end AI system has achieved drug discovery at a fraction of the cost and time.

Many companies started experimenting with end-to-end AI systems around 2019-2020, making this a relatively new but fast-growing field.

Insilico uses AI to generate novel molecular structures to target specific biological mechanisms.

Insilico Medicine is releasing a Pharma.AI update on the 14th, signaling a major advancement in AI for drug discovery.

AI will democratize and de-risk drug discovery, allowing new entrants, including less experienced countries, to innovate in biotech.

In the next 20 years, Alex predicts that AI systems will enable personalized drug discovery, tailoring drugs for individual patients.

The future of AI in biotech is faster and more promising than many people currently believe, with transformative impacts on healthcare.

Transcripts

play00:07

I do not call any place home.

play00:11

I travel one hundred percent of my time.

play00:14

I don't own an apartment.

play00:17

I usually sleep about four or five hours a day.

play00:23

I don't have a family and Insilico is my life.

play00:33

My name is Alex Zhavoronkov

play00:34

and I am the founder and CEO of a company called Insilico Medicine.

play00:39

I specialise in AI-powered drug discovery,

play00:43

so developing artificial intelligence solutions

play00:46

to accelerate the way drugs are discovered and developed.

play00:52

Around 2013, 2014, deep neural networks became so powerful

play00:57

that they started outperforming humans in image recognition

play01:01

text recognition and many other human tasks.

play01:06

The public perceives AI as something magical

play01:09

and something that may replace human intelligence in daily tasks.

play01:16

It's actually not far from the truth,

play01:19

but what AI really is is mathematics and neuroscience.

play01:24

To me, AI represents hope.

play01:31

To us, AI is a tool which allows us to go into drug discovery much faster,

play01:37

go after much more challenging biological and chemical problems

play01:41

with higher confidence,

play01:43

and reduce the risk in drug discovery

play01:46

and also increase the probability of success in clinical trials.

play01:51

The pharmaceutical industry is extremely risky.

play01:54

It takes decades to develop a new drug and bring it to the market,

play01:58

and you fail more often than you succeed.

play02:06

So, you really need to be the true mixed martial arts fighter in pharma

play02:10

in order to succeed.

play02:13

What we have done at Insilico so far,

play02:16

we've demonstrated that we can go from disease hypothesis

play02:19

all the way into clinical trials,

play02:21

with very high probability of success.

play02:24

So, we actually validated our AI utilising our own therapeutic programmes,

play02:29

going end-to-end from the very risky part

play02:32

where we discover and uncover and validate novel targets,

play02:35

all the way to generating our own novel chemistry

play02:38

with the desired properties.

play02:40

And we've managed to do it at a fraction of the cost and time.

play02:47

Most of the companies in our field

play02:49

started experimenting with end-to-end systems around 2019, 2020,

play02:55

so it has really been only two or three years.

play03:01

So, we started utilising the same technology

play03:04

to generate novel molecular structures with the desired molecular properties

play03:08

to prosecute those biological targets that we identify using a different form of AI.

play03:12

On the 14th, we are releasing our Pharma.AI update.

play03:20

The future is much better than you think, and also much faster.

play03:28

AI will definitely de-risk, democratise and enable new entrants,

play03:36

including younger countries who have never had any experience in drug discovery,

play03:43

to get into biotech, to join the community and to start innovating from the get-go.

play03:49

I believe that in 20 years,

play03:52

we are likely to see AI systems going after personalised drug discovery,

play03:58

where drugs will be tailored towards a specific individual.

Rate This

5.0 / 5 (0 votes)

الوسوم ذات الصلة
AI drug discoverybiotech innovationpharma AIpersonalized medicineclinical trialsdeep learningneural networkstherapeutic AIInsilico Medicinebiological targets
هل تحتاج إلى تلخيص باللغة الإنجليزية؟