In the vital search for new therapeutics, whether small molecules or biologics, there is great value in any approach that can focus experiment or deliver breakthrough insights. Machine learning is one such technology, with the potential to extract more from existing data, learning vital relationships that enable it to predict properties and accurately assess the probability of likely outcomes. Investigate candidate compounds, mine data on drug-target interactions, identify drugs that might be repurposed, and guide experimental programmes. Adoption of machine learning can be constrained by the need to handle large datasets that are often sparse and noisy. The Alchemite™ technology can master such real-world drug discovery data.
Analysing a large pharmaceutical dataset with Takeda
A peer-reviewed study has been published in Applied AI Letters. Working with Takeda’s proprietary global dataset, the project team demonstrated the ability of machine learning imputation to reduce cost and improve success rates of drug discovery.
Optimising kinase profiling programmes
A project at Genentech applied the Alchemite™ method to optimise kinase profiling programmes. The data set studied consisted of approximately 650 kinases and 10,000 compounds. Alchemite™ significantly outperformed biological fingerprint similarity and quantitative structure-activity relationship (QSAR) approaches, including multi-target deep learning. The insights developed enabled active learning, assay prioritisation, and virtual screening.
Finding a novel antimalarial compound
In work published in the Journal of Medicinal Chemistry, Intellegens participated in a global challenge organised by the Open Source Malaria consortium to design new antimalarial compounds with a novel mechanism of action, despite the sparsity of available experimental bioactivity data. The active compound identified by Alchemite™ deep learning show to have good potency when synthesised and tested. Predictions from Alchemite™ outperformed alternative machine learning approaches assessed by the consortium.
Intellegens partners with Optibrium to embed the Alchemite™ deep learning algorithms within Cerella™ – a software platform for active learning in drug discovery. Cerella™ users can gain more value from compound data, capturing complex relationships between structures, activities and other properties on a scale that cannot be achieved using conventional tools and supporting chemical space exploration beyond conventional QSAR models. Cerella™, powered by Alchemite™, increases confidence in decision making and lead optimisation and enables you to reduce costs and accelerate discovery cycles by targeting experimental resources more effectively.
Intellegens publications on drug discovery
Find out more about the science behind the Alchemite™ methodology for drug discovery:
Alchemite™ for drug discovery
Alchemite™ technology enables you to:
- Fill gaps in your compound and biologics data, avoiding missed opportunities
- Capture complex relationships between structures, activities, and other properties
- Use accurate uncertainty estimates to prioritise therapeutics most likely to succeed
- Focus your experimental resources on the most valuable measurements
- Explore chemical space beyond conventional QSAR models
- Standardise and share data and models while ensuring rigorous security for your IP