Biological drug target discovery from transcriptomic data - building a digital twin of disease using machine learning - a breast cancer case study with Prof Graham Ball

  • Date: 1 June 2022
  • Time: 13:00 pm
  • Cost: Free
  • Venue: Online
Join the Teams event on Wednesday 1 June at 1pm
Graham Ball

Join ARU's Medical Technology Research Centre (MTRC) for an online seminar from Prof Graham Ball on using machine learning to better understand and treat breast cancer.

Breast Cancer is a prevalent disease. In 2020, there were 2.3 million women diagnosed with breast cancer and 685 000 deaths globally (WHO).

The tumour suppressor gene TP53 has a unique role in directing the fate of the cell in tumorigenesis. It has diverse functions in cell cycle regulation and apoptosis. It acts as a central player and as a transcriptional factor for a wide range of other genes, which collectively form the TP53 pathway.

Inactivation and somatic mutation of TP53 are common in breast cancer. Around 50% of human tumours harbouring alteration in the TP53 gene. Here we adopt a machine learning-based systems biology approach to the modelling of the P53 pathway.

The aim is to identify unique gene expression signatures that associate with the TP53 pathway using an evidence-based approach across the whole transcriptome. This is achieved by using a combination of artificial neural network (ANN) based data mining and ANN-based network inference using the P53 pathway as a framework for the interrogation of transcriptomic breast cancer data (TCGA, focusing on Triple-negative phenotypes). Following this, we use network inference approaches to model the pathway and identify molecular drivers at the systems level.

Through these approaches, we build a digital twin of the disease at the molecular level. Interrogation of this identifies novel drivers of the pathway in different molecular subtypes. The pathway drivers can also be used as biological targets for drug discovery.

Prof Graham Ball is the new director of MTRC and Founder-CSO of Intelligent Omics. He specialises in the application of innovative artificial intelligence and machine learning methods to the analysis of complex data in the biomolecular, biomedical, and pharmacological domains.

Prof Ball has led the development and validation of bioinformatics algorithms using machine learning to mine molecular data for the last 20 years. He has focused on applying these approaches to public data repositories leveraging actionable and translational features from the data. He has published over 200 journal papers and seven patents in this area.

After a PhD (UN-funded) modelling environmental systems, in 2000 Prof Ball shifted his focus to analysis of proteomic and genomic data, searching for proteins and genes associated with cancer. His current research focuses on innovative computational methods that allow the identification of optimized biomarker panels, molecular systems of disease and druggable biology.

Contact MTRC@aru.ac.uk for more information.

  • Date: 1 June 2022
  • Time: 13:00 pm
  • Cost: Free
  • Venue: Online
Join the Teams event on Wednesday 1 June at 1pm