Queen Mary University of London seeks to advance AI for drug discovery

Queen Mary University of London has agreed a new collaborative research project with AstraZeneca to advance AI for drug discovery. The research will be led by Professor Greg Slabaugh, Director of Queen Mary's cross-disciplinary Digital Environment Research Institute.

The researchers will develop and refine algorithms for multimodal data integration. These algorithms integrate various types of omic data-such as RNA, protein, and imaging-from pharmacologically or genetically modified cells. The goal is to expedite small compound drug discovery by identifying the most informative omics technologies, creating reference datasets, and maximising value from AstraZeneca's ongoing efforts in medicinal chemistry.

To do this, Queen Mary will apply their existing machine learning algorithms and academic expertise to AstraZeneca's multi-omic datasets. They will refine and develop them to create new omics analytical capabilities to look for ways to streamline the drug discovery process, making it faster and more cost-effective.

Queen Mary's Professor Greg Slabaugh comments:

"Artificial intelligence has significant potential to bring new treatments to more patients faster - so we're proud to be working with AstraZeneca to develop new machine learning platforms to accelerate drug discovery. By combining our academic expertise with industry data and biological knowledge, we hope to make a transformative impact for patients and the entire field."

It typically takes over ten years and more than $1bn to bring a new drug to market, so accelerating the process using AI has great potential to reduce time and costs. The Queen Mary-AstraZeneca AI tools will be developed to be useful across multiple applications and stages of the R&D pipeline.

The project starts in January 2025 and will run for 30 months. The funding is from AstraZeneca, UKRI, and Queen Mary.

The team are currently hiring for an associate researcher. If you have a keen interest in life sciences, experience of interdisciplinary research, and a PhD in computer science, bioinformatics or similar, then we'd love to hear from you. Find out more and apply.

Multi-omics modalities: Different types of biological data that are studied together to understand how living systems function. Each "omics" modality focuses on a specific type of biological information: Genomics (the DNA sequence and gene structure), Transcriptomics (RNA which characterise how genes are expressed), Proteomics (protein molecules which carry out most biological functions), and images. By combining data from multiple omics, scientists can better understand the complexity of biological processes, diseases, or treatments.

Multimodal data integration algorithm: A computer method used to combine and analyse different types (modalities) of data. For example, if you're studying a disease, you might have different types of data like genetic information (genomics), protein levels (proteomics), and medical images. Each type of data gives some insight, but to fully understand the disease, you need to look at all the data together. A multimodal data integration algorithm helps merge and analyse all these different data types to find patterns or insights that wouldn't be clear from just one type of data alone.

Learn more about industry partnerships with Queen Mary.