Biotech start-ups tap into the power of AI for drug discovery and protein design

To its humanoid subordinates, artificial intelligence (AI) already has become a life saver for students who have favoured partying over essay writing, or for harried journalists with a looming deadline.

AI is also emerging as a life saver in a more literal sense: as a tool to expedite the notoriously slow path of drug development and to advance personalised medicine.

It’s well known that a blockbuster drug usually takes more than 10 years to get to market and billions of dollars to develop.

According to the Albert Einstein College of Medicine – named in honour of one of the world’s greatest human brainiacs – there’s only a 0.1% chance of a blockbuster drug (defined as having annual sales of US$1 billion or more) ever getting to market.

Even lesser drugs try the patience and the purse strings of investors.

In a rare win for an Australian biotech, the US Food and Drug Administration (FDA) this week approved the ASX-listed Neuren’s (ASX: NEU) drug to treat a rare neurological diseases – a therapy that took more than $200 million and almost two decades to develop.

Advantages of AI

But expediting the drug development time line and lowering the cost, machine learning offers huge advantages from early discovery to commercialisation.

At the early stage, AI can identify new drug targets and molecules that are more likely to be effective against those targets.  AI can analyse vast amounts of data from multiple sources such as biological databases, scientific literature, and clinical trial data.

In clinical trial design, AI can identify the patients most likely to respond to the proposed drug, as well as potential safety issues before they become a problem.

As its name suggests, personalised medicine is the tailoring of drug treatments based on the individual patient’s genetics and medical history.

Once again, AI can identify which patients are likely to respond to a certain treatment and reduce the risk of nasty side effects.

AI start-ups tap the trends

While biotech valuations have sagged like last night’s souffle, an array of biotech start-ups are focused on ‘generative’ AI that gets better as it crunches more data.

For example, Dutch start-up Cradle recently raised €5.5 million (A$8.85 million) with the remit of using AI to design and improve proteins. Such synthetic biology is relevant not just for medicines, but generating alternatives to food, clothing, materials and chemicals.

The cofounder of mental illness-focussed AI start-up Thymia, Dr Stefano Goria, concurs the opportunities are massive.

“Drug discovery, antibody discovery, and protein discovery are all fantastic applications of the tech,” Dr Goria told Business Insider.

(Dr Goria, by the way has a PhD in theoretical physics so he’s still much smarter than the average 8-bit Commodore C64).

The diagnostics revolution

AI is also exerting its powerful influence in diagnostic imaging: the everyday use of tools such as magnetic resonance imaging, computer tomography (CT) scans and X-rays to detect tumours and other maladies.

For its part, the FDA has already approved hundreds of algorithms to improve the efficacy of these tools, for uses including cardiology, musculoskeletal and neurological.

“The industry is now progressing into revenue-generation and becoming increasingly competitive,” says broker Goldman Sachs in a recent report.

“Based on efficiency improvements cited in the clinical studies we examined, and our conversations with industry practitioners, we see the potential for AI to disrupt 30% to 40% of time spent on image interpretation by radiologists.”

Return of the woolly mammoth?

Cradle founder Stef van Grieken describes AI as “decoding DNA so humans can do what we want with it.”

Given that, one can only ponder the imminent return of the Tasmanian tiger or even the woolly mammoth. But the white-coated scientists can be comforted that as with all forms of computing, AI won’t circumvent human endeavour altogether.

McKinsey & Co partner and biotech guru Alex Devereson believes drugs might be developed in one-tenth of the time they are now and they will be more applicable to untreatable diseases.

“Fundamentally, we will have life-changing, game-changing drugs – on a scale and at a pace that we’ve never seen before – getting to the right patient at the right time,” he says.