Putting AI to work in longevity research: profile of Alex Zhavoronkov

By Calum Chace | 13 September 2021
Calum Chace

(Photo: Dreamstime.com)

A man on a mission

Alex Zhavoronkov is a man on a mission. He is extremely focused – some might say obsessive. He works 18 hours a day 365 days a year. He takes no holidays, and has no plans to start a family – unless he meets a potential partner who is at least as focused as himself.

The subject of this relentless attention is the aging process – how to understand it, arrest it, and reverse it. In 2013, Zhavoronkov published a book called “The Ageless Generation” which argued that everyone over about 30 years old is effectively sick, and that curing this sickness is both possible, and a moral and economic necessity. It is an economic necessity, he argues, because the world will soon be unable to support a senior population which cannot work, and which makes enormous demands on healthcare budgets as it suffers the symptoms of aging. In 2013, the US national debt reached $10 trillion. Today, less than a decade later, it is approaching $30 trillion, and it is projected to exceed $50 trillion in 2026. That is well over half the GDP of the entire world today.

Today, Zhavoronkov’s day job is running Insilico, a drug discovery company in Hong Kong that he founded in 2014, which uses artificial intelligence to identify novel targets and novel molecules for pharmaceutical companies. Insilico Medicine recently span off an aging biomarker and app business called Deep Longevity, and sold it to Regent Pacific Group, Jim Mellon’s investment firm specialising in healthcare and late‐stage life sciences. Zhavoronkov has taken on the role as Chief Longevity Officer at Deep Longevity. He spends only 10% of his free time on this role, but because he works such long hours, he observes dryly that this is more time than most people spend on their full-time jobs.

The vice of sleep deprivation

Obviously, he has no time for drink, drugs, or gambling, and he jokes that his only vice is missing out on sleep. As an expert on aging, he fully understands the gravity of this vice, but he believes his work to be important and urgent enough to justify it.

The focus is paying dividends. A potentially powerful new drug developed by Insilico reached pre-clinical trial stage this February. Thanks to the company’s innovative use of sophisticated AI, the development process took 18 months and cost just $2 million, compared to several years and hundreds of millions of dollars in a traditional pharma company.

The disease being addressed is Idiopathic Pulmonary Fibrosis (IPF). Its causes are unknown (hence “idiopathic”), and it stiffens the lung tissues of older people, and can eventually kill them. Insilico used natural language processing AI to trawl through massive medical data sets, looking for a protein which could be causing the problem. This protein is called the target. Once the target was identified, Insilico used a type of AI called a Generative Adversarial Network (GAN) to identify a molecule which could become an effective drug against the disease. GANs pit two AIs against each other: one generating suggestions, and the other critiquing those suggestions in an evolutionary process.

Using AI to improve drug development

This is a stunning achievement, and not only for the five million or so people who are affected by IPF each year. It proves the potential for modern AI techniques to dramatically improve the drug discovery process, a breathtakingly expensive and time-consuming process in which 90% of all drug candidates fail.

The IPF announcement was no one-off. On 4th August, a couple of days before our conversation for this article, Insilico announced that it had repeated the performance with a molecule to treat kidney fibrosis. About 10% of the world’s population suffers from kidney disease, and fibrosis is a common cause. If Zhavoronkov achieves nothing else in his career, these breakthroughs will be enough to earn him a place in the medical history books. Given that he has just turned 42, there are likely to be many more.

Fierce ambition

Zhavoronkov has wanted to pursue a career in tackling aging since his boyhood in Latvia, where he learned to speak Russian as well as Latvian. He decided to make some money first, so he travelled to Canada, to study IT and commerce at Ontario’s prestigious Queen’s University. Zhavoronkov’s fierce ambition was already evident: his tutors did not know that he was taking two degrees simultaneously, and his final year workload was punishing. This was his first period of extended sleep deprivation. As if studying for two degrees was not enough, he was also working with a friend on a startup.

That startup never launched, but he soon joined a young company called ATI, a rival to Nvidia in Graphics Processing Units (GPUs), the chips that rose to prominence thanks to video games, and which now power deep learning AI systems. He bought some ATI stock, and when it was sold to AMD in 2006 he had enough money to take a few years off.

He was already taking a masters degree at John Hopkins University, and here he met Charles Cantor, a director of the Human Genome Project. Cantor became a mentor, and in 2009 he engaged Zhavoronkov as a consultant to establish a JV in Russia for Sequenom, a Californian company working in non-invasive pre-natal testing.

Deep learning

During the next few years Zhavoronkov pursued his goal of working in anti-aging science, and also became convinced that AI would be an important tool for the field. He was working with talented Russian engineers and former colleagues in the GPU manufacturing community to explore the potential of deep learning, which caused the Big Bang in AI in 2012, when Geoff Hinton finally got an algorithm called backpropagation to work, and achieved a famous breakthrough in image recognition.

Observing that more and more aspects of the computer industry were being outsourced to China, Zhavoronkov decided to move to Hong Kong. He remains bullish about China’s future contributions to the field of anti-aging. An indicator of the scale of its ambitions is China Medical City, a medical research centre about half the size of Manhattan, being built in Taizhou, a city about 150 miles northwest of Shanghai.

“The Ageless Generation”

When he wrote “The Ageless Generation” in 2013, Zhavoronkov was still feeling his way into the anti-aging space. He was cautiously optimistic that progress would be made fast enough to head off the budgetary crunch which he sees as otherwise inevitable. Nearly a decade later, he is a well-respected leading figure in the space, but he is more daunted by the task. He continues to see the exponential growth in the performance of the technologies that we need to solve the problems, but since the biology turns out to be even more complicated than he previously thought, he is less confident of timely success.

The pharmaceutical industry badly needs re-inventing, to make it faster and less costly, and to make it focus directly on the basic biology of aging, not just the diseases like cancer, heart disease and dementia which are caused by aging. One alumnus of Queen’s University (Elon Musk) forced the automotive industry to re-invent itself, and switch from internal combustion engines to electric ones. Maybe another of its alumni will do the same for pharmaceuticals.

Reprinted with permission from the author.

Calum Chace is an English writer and speaker, focusing on the likely future impact of Artificial Intelligence on people and societies. He became a full-time writer in 2012, after a 30-year career in business. He is the author of Surviving AI, The Economic Singularity, and the philosophical science fiction novel Pandora’s Brain.

Calum Chace – ‘We cannot stop the progress of AI’

Anticipating 2025: Calum Chace

Artificial Intelligence-based Biomarkers of Ageing | Dr Alex Zhavoronkov

Be sure to ‘like’ us on Facebook


Please enter your comment!
Please enter your name here