eCube Harnesses Machine Learning to Identify CNS Compounds and Predict Their Therapeutic Applications for Neuropsychiatric Disorders
PsychoGenics Inc., a provider of AI-enabled phenotypic drug discovery and preclinical CRO services, announced today that it had launched eCube, a novel pharmaco-electroencephalography (“pharmacoEEG”) platform to support phenotypic drug discovery, which, unlike traditional target-based drug discovery, can identify novel treatments with multiple targets. Similar to the Company’s AI-enabled, behaviour-based platform, SmartCube, eCube harnesses machine learning to identify CNS penetrant compounds and predict their therapeutic applications for neuropsychiatric disorders by comparing their EEG profiles to an extensive pharmacoEEG database comprised of hundreds of doses of CNS-active reference drugs.
eCube measures drug-induced changes to EEG spectra data and utilises machine learning classifiers trained on reference drugs to predict the efficacy of novel compounds at the drug class (primary therapeutic indication) and subclass (primary mechanism of action) levels. In addition to predicting therapeutic applications for novel compounds, eCube can deliver translational biomarkers, providing an objective way to assess CNS activity, which is particularly important for compounds with unknown or complex mechanisms of action.
“The eCube platform joins PsychoGenics’ innovative class of AI-enabled phenotypic platforms, including SmartCube, which has helped advance numerous compounds to clinical trials, including Ulotaront. Discovered in partnership with Sunovion, Ulotaront is now in Phase 3 for schizophrenia, as well as depression and generalised anxiety disorder,” said Emer Leahy, PhD, CEO, PsychoGenics. “eCube is another important discovery tool helping bring novel treatments to patients suffering from severely disabling conditions.”
“The introduction of eCube provides a truly unique translational tool for target-agnostic compound screening and lead generation. Our reference EEG databases and analytical machine learning tools allow us to predict therapeutic efficacy and quantify the similarity of EEG spectra to other CNS drugs,” added Dani Brunner, PhD, Chief Innovation Officer, PsychoGenics. “This powerful platform can confirm the therapeutic predictions from our behaviour-based platforms, identify neuronal circuitry and provide translational biomarkers to support clinical development.”
