My research group is primarily interested in the development and application of statistical/machine learning methods in different biological contexts. Examples include:
1. Risk stratification in transplant patients
Current projects include the development of a PSO-optimsed ensemble classifier to provide better predictions of graft survival in kidney transplant patients. This work is focused on sub-clinical detection of antibody-mediated rejection (AMR), a leading cause of graft failure, and is being carried out in collaboration with the kidney transplant program at the Albert Einstein College of Medicine in New York.
2. Cancer Genomics
Current projects include: i) Understanding the role of the bone marrow microenvironment in AML progression and drug resistance, and ii) the development of graph-based machine learning methods to help elucidate the role of circRNAs in prostate cancer drug response. This work is being carried out in collaboration with colleagues here at NUIG as well as at St James’s hospital/TCD.
(Zhang et al. npj Precision Onc, 2017)
This work stems from my affiliation with NICOG – the Centre for Neuroimaging and Cognitive Genomics. Current projects include: i) the development of novel machine learning methods for polygenic risk scoring in genome-wide association studies related to schizophrenia/cognitive function, and ii) the use of convolutional neural networks (CNNs) to identify image-derived phenotypes in MRI scans of bipolar disorder patients for improved intepretability in GWAS studies.