PsychoGenics is using AI to rapidly screen for psychotic treatments

Rather than targeting specific genes associated with the disorder, the company analyses patterns of behaviour and the output of the brain to trial treatments. PsychoGenics’ lead neuroscientist, Dr Emer Leahy, describes SmartCube as the company’s workhorse for drug discovery.

Candidate drugs are injected into mice within a hyper-controlled environment where their behaviour is monitored and autonomously tracked over the course of an hour.

‘SmartCube is essentially a box that challenges mice in a variety of ways, while multiple cameras give us a 3D picture of everything that mouse is doing,’ explains Leahy.

‘We’re collecting a couple of million data points from SmartCube, so we’re using machine learning to extract patterns of behaviour and make predictions about therapeutic utility,’ she says.

The use of mice is ethically questionable for obvious reasons, but leveraging the power of AI has undoubtedly improved the results of testing. PsychoGenics had to test around just 300 compounds, compared to the average of 2,500, for a promising treatment to emerge called ‘Ulotaront’.

Found to treat the positive symptoms of schizophrenia – i.e. hallucinations, delusions, and hearing voices – what distinguishes Ulotaront is its reported lack of secondary symptoms.

Feelings of apathy and social withdrawal are usually part and parcel of psychotic medications, but the team documents a significant reduction on both fronts with their product of clinical AI. If correct, Ulotaront could be a game changer.

Due to be investigated further in upcoming clinical trials in Japan and China, there is optimism that the drug could be up for FDA approval in the near future. If that were to happen, AI could transform the way we trial medications way faster than projected.

Rounding off on the potential of AI in clinical trials, Leahy declares: ‘We’re in a new age.’