We collaborate with the Faculty of IT in the area of pharmaco-informatics. Each of the projects below are undertaken in a research strength of the Faculty of Pharmacy and Pharmaceutical Sciences, with joint supervision from academics in the Faculty of IT.
Predictive tools that enable the design of orally bioavailable high molecular weight (MW) drugs are keenly sought and have become a key priority for many academic and industry research teams. However, computational models that guide drug design for high MW (> 500 Dalton) molecules are currently not sufficiently predictive for routine use in drug design settings. This project seeks to address this limitation and will design new AI models of oral bioavailability.
This project will suit students with a background in artificial intelligence, computational modelling, drug design or pharmaco-informatics.
This project will be a collaboration between the Faculty of Pharmacy and Pharmaceutical Sciences and the Faculty of Information Technology, and supervised by Nick Barlow (FPPS), Buser Say (FIT) and David Chalmers (FPPS).
Undertake a placement of a defined period in an external institution, either domestic or international, gaining experience in a different research environment that will augment your PhD.
This PhD project will include working with a large Australian linked administrative health care data focused on chronic diseases such as cardiovascular diseases, diabetes and dementia. The project will involve applying the Observational Medical Outcomes Partnership (OMOP) Common Data Model to the Australian dataset, and validation of this application in the Australian setting for medication effectiveness and safety research in chronic diseases.
Undertake a placement of a defined period in an external institution, either domestic or international, gaining experience in a different research environment that will augment your PhD.
Project background:
Heart failure remains a leading cause of morbidity and mortality worldwide. The recent availability of whole exome sequencing of hundreds of thousands of individuals allows genome-wide association studies (GWAS) to identify associations of genotypes with phenotypes. Recent GWAS studies have identified several G protein-coupled receptors (GPCRs) associated with heart failure and/or related cardiovascular disease that represent novel therapeutic targets. GPCRs control critical physiological functions and are known to be tractable drug targets, as evidenced by over one-third of FDA-approved small molecule medicines targeting this class of receptors.
Project aims:
1) To apply a Bayesian framework to integrate multi-omics data and gene networks to further explore associated GPCRs from GWAS data
2) To understand the influence of disease-associated GPCR mutations on protein expression, localisation and function
3) To apply a physicochemical graph neural network alongside compound screening to identify new drug leads targeting GPCRs associated with cardiovascular disease.
Techniques applied include data science and artificial intelligence approaches, computational biology, molecular biology, signalling assays, cell culture, and high-end fluorescent microscopy.
Undertake a placement of a defined period in an external institution, either domestic or international, gaining experience in a different research environment that will augment your PhD.
Dementia, affecting 57 million people globally in 2019 with estimates rising to ~150 million by 2050, is particularly prevalent in women. Postmenopausal hypertensive women undergo the fastest brain health decline, potentially due to reduced G protein-coupled estrogen receptor (GPER) signalling from lower postmenopausal estrogen levels. While GPER activation could combat dementia, current GPER agonists often suffer from poor selectivity and solubility, stressing the need for new, targeted treatments. This project aims to leverage artificial intelligence and computational biology to discover novel selective GPER agonists.
Aim 1. Reveal atomic mechanisms of GPER selective agonism through deep learning approaches.
Aim 1.1. To predict GPER networks using AI-based deep learning approaches applied to GaMD simulations.
Aim 1.2. Pharmacological evaluation of predicted selectivity networks.
Aim 2. Design a novel selective GPER agonist.
Aim 3. Evaluate the efficacy of GPER agonists to reduce dementia-like changes in vivo.
Undertake a placement of a defined period in an external institution, either domestic or international, gaining experience in a different research environment that will augment your PhD.
Cardiopulmonary disease is the medical term used to describe a range of serious disorders that affect the heart (“cardio-”) and lungs (“-pulmonary”), such as pulmonary hypertension, chronic respiratory disease, and heart failure. There is urgent unmet need to discover novel druggable targets. G protein-coupled receptors (GPCRs) are the most intensively studied drug targets (~34% of all drugs approved by the FDA), largely due to their substantial involvement in human pathophysiology and their pharmacological tractability.
Machine learning-based artificial intelligence (AI) offers a powerful approach to significantly augment protein engineering by constructing AI models capable of making predictions of the sequence-structure-function relationships of GPCRs. Underpinned by our internationally leading research in bioinformatics, machine learning and drug development, in this PhD project, we will be the first to employ cutting-edge generative AI, pharmacology and computational biology to discover novel drug targets for GPCR-based disease and treatment.
We will develop a novel computational framework based on the generative AI models to design ‘allosteric’ inhibitors targeting GPCRs in silico, perform benchmarking experiments to assess the utility of the framework for inhibitor discovery for different types targets and experimentally validate the designed inhibitors. This interdisciplinary project will provide a unique opportunity to train a young and talented PhD student at the interface of AI, computational biology, drug discovery and medicinal chemistry.
Undertake a placement of a defined period in an external institution, either domestic or international, gaining experience in a different research environment that will augment your PhD.
DNA encoded libraries (DEL) typically contain billions of compounds that are screened to identify those that bind to a specific target. Analysis of the output of DEL screening usually considers only those compounds that are enriched in the screening process. This means that >99.99% of the data that is generated in the screen i.e. the compounds that do not bind, is ignored in the analysis. The "best" of these hits are then re-synthesised off-DNA and re-screened to validate binding. More recently, machine learning and artificial intelligence approaches have been applied to the analysis of DEL screening data. The aim is to use all of the data that are generated in the screen to develop a pharmacophore model of the binding site. This can then be used to select commercially available compounds to test the binding model. This accelerates the process of hit identification and reduces the cost of generating hits, since no synthesis is required to validate the binding hypothesis.
Undertake a placement of a defined period in an external institution, either domestic or international, gaining experience in a different research environment that will augment your PhD.