2022 Collaborations

  • Unlocking diverse herbarium specimen associated data to accelerate biodiversity and evolutionary research

    This collaboration will increase access to herbarium-derived biodiversity data by creating tools to capture, analyse, and share herbarium specimen associated data.

    Biodiversity data, such as those that accompany museum and herbarium specimens, comprise a massive global data resource. This data can be used for rapid biodiversity analyses and to inform conservation and policy decision making. To do this it needs to be accessible within an integrated biodiversity infrastructure, that retains links among spatial, trait, and evolutionary data for taxa.

    This project will provide tools to support efficient mobilisation of high-quality specimen associated primary (e.g. taxon, location) and secondary (e.g. molecular) biodiversity data. It will generate a pipeline for collation and phylogenetic analyses of molecular data generated by University of Melbourne researchers and students. It will also create an online phylogenetic browser to enable exploration of native Australian plant and algal evolutionary trees. This will be hosted on the University of Melbourne Herbarium Collection Online for access by students, researchers, and the global public.

    Chief Investigator: Jo Birch, School of Biosciences, Faculty of Science  

    MDAP Collaboration Leads: Robert Turnbull and Karen Thompson

    Collaborators: MDAP, the Faculty of Science, and Royal Botanic Gardens Victoria

  • Screening whole genome sequencing datasets for the Eukaryome of globally distributed kelp species

    This collaboration aims to screen whole genome sequencing datasets of globally distributed kelp species for host-associated eukaryotic taxa, filling a considerable gap in kelp holobiome biology.

    Bacterial diversity of seaweed microbiomes has received considerable attention over the years. But far less is known about their eukaryotic component such as algae, fungi, and animals living in/on organisms – collectively called the Eukaryome. We need new diagnostic frameworks for characterising eukaryotic microbial diversity. This will allow us to meet increasing industry demands for certified seaweed products, and better understand ecologically significant kelp forest communities.

    The collaboration will help establish a robust workflow for future genomic analyses of entire microbial communities. It will employ state-of-the art machine learning algorithms to classify sequences and enhance knowledge about kelp holobiomes.

    Chief Investigator: Trevor Bringloe, School of Biosciences, Faculty of Science          

    MDAP Collaboration Lead: Noel Faux 

    Collaborators: MDAP, CSIRO, Sungkyunkwan University (South Korea), the Athlone Institute of Technology (Ireland), the National University of Ireland Galway, and the University of Victoria (Canada)

  • Infectious disease genomics: from database to phylodynamics

    The ongoing SARS-CoV-2 pandemic has seen an unprecedented amount of genome data generated, with over 2 million complete genomes in the GISAID platform. These data have been pivotal to track the evolution of the virus on a global scale and to detect mutations of epidemiological importance, such as those in variants of concern.

    Many important biological questions could be addressed with these data. However, there can be issues assessing data quality and suitability. The main limitation in accessing sequence data from public databases is that many genomes have substantial stretches of missing data and are thus not sufficiently reliable to infer mutational events. Others can lack important metadata, such as travel history or vaccination status. 

    This project has two key aims: 

    1. Develop a system where one can select thousands of sequences that meet certain criteria for quality control or the presence of metadata that are fit-for-purpose for epidemiological questions
    2. Develop tools to specify Bayesian hierarchical models in an intuitive format that will use the above fit-for-purpose data. 

    Chief Investigator: Sebastian Duchene, Melbourne School of Biomedical Sciences, Faculty of Medicine, Dentistry & Health Sciences 

    MDAP Collaboration Leads: Simon Mutch & Priyanka Pillai 

  • FATCAT: a focused pipeline to assess functional changes in coral-associated bacterial communities

    This project will develop a Functional Annotation Tool for Coral-Associated bacterial Taxa (FATCAT) to better predict coral-associated bacterial functional profiles. FATCAT will expand beyond traditional GTDB and NCBI databases with the addition of coral-associated bacterial genomes.

    There have been encouraging examples in the literature of efforts to make environment-specific databases. For example: CowPI, a functional inference tool specific to the rumen microbiome, and FAPROTAX, designed to assess microbial communities from aquatic samples.

    We plan to update FATCAT in sync with the release of new coral-associated bacterial genomes. This will make FATCAT the standard to accurately predict bacterial functions from 16S rRNA gene based coral microbiome studies.  

    Chief Investigator: Ashley Dungan, School of Biosciences, Faculty of Science

    MDAP Collaboration Lead: Bobbie Shaban 

  • Machine learning and patterns of primary care utilisation in cancer diagnosis and outcomes

    This project seeks to apply novel analytical and machine learning methods to the large-scale linked data sources. This will generate new hypotheses and further characterise how people with cancer engage with primary care services.

    The project aims to identify patterns in various aspects of primary cancer attendances prior to a definitive cancer diagnosis. These could include:

    • prescribing
    • test requests and results
    • semi-coded fields such as ‘reason for encounter’
    • co-morbidities or other conditions captured in primary care management systems. 

    The project will identify a specific cancer type (i.e. Upper Gastrointestinal) with sufficient linkages between hospital diagnosis and primary care data.  It will also apply tools and algorithms to detect patterns present within specific timeframes of diagnosis or treatment.

    Chief Investigator: Jon Emery, Centre for Cancer Research, Faculty of Medicine, Dentistry & Health Sciences

    MDAP Collaboration Lead: Zaher Joukhadar 

  • Artificial intelligence-based image enhancement and segmentation

    This project aims to convert low-resolution CT images of large samples to super-resolution by using artificial intelligence (AI)-based image processing algorithms. AI algorithms are supposed to be used to split connected particles in CT images. Then, the particle volume, particle surface area and interparticle contact areas will be computed from the enhanced super-resolution CT images. The values will be compared with their counterparts derived from the low-resolution CT images before enhancement.  

    Chief Investigator: Wenbin Fei, School of Electrical, Mechanical and Infrastructure Engineering, Faculty of Engineering and Information Technology      

    MDAP Collaboration Lead: Jonathan Garber

  • Pipeline for the refinement of predicted protein structures and evaluation of their accuracy

    Most 3D crystal structures for proteins of eukaryotes come from human and model organisms such as mouse, fruit fly and yeast. Until recently, there has been no method to acquire 3D protein structure for other organism such as parasitic worms and other pathogens.

    The program AlphaFold, developed by DeepMind and published in July 2021, is about to "close the gap" between the quality of predicted proteins structure to those acquired from crystal structures. However, despite the progress, many predicted protein structures do not exhibit sufficient accuracy for applications such as drug design. To address this we need to refine structures using molecular dynamics (MD) simulations.

    This project will develop a software pipeline for the refinement of predicted protein structures and the evaluation of their accuracy.

    Chief Investigator: Robin B. Gasser, Melbourne Veterinary School (Graduate School), Faculty of Veterinary and Agricultural Sciences

    MDAP Collaboration Leads: Bobbie Shaban & Edoardo Tescari 

    The project is supported by Australian Research Council Linkage Project (LP180101085) – ‘Illuminating genomic dark matter to develop new interventions for parasites’

  • Tracking changes in conservative political rhetoric over time

    This project seeks to develop tools to investigate changes in liberal and conservative political rhetoric in Australia over time. Researchers and the public will be able to undertake their own temporal and linguistic analysis using unique primary sources, such as raw text files, transcriptions, and pdf images.

    The collaborative project will also develop online natural language analysis tools, using Python, that will make visualisation of the data possible via an online interface. This project is designed to be scalable, allowing for the input of more primary sources in the future. This will enable the University to partner with cultural institutions such as the National Library of Australia. The tools will be used in the Faculty of Arts for teaching purposes. For example, embedding analytics and research data management exercises in selected subjects. It will also provide an ongoing resource for postgraduate and early career researchers.

    Chief Investigator: David Goodman, Digital Studio, School of Historical and Philosophical Studies, Faculty of Arts

    MDAP Collaboration Lead: Mel Mistica 

  • Using the capability of deep learning to explore the predictive power of 180,000 eye images for dementia outcomes

    This project will examine the ability of non-invasive eye imaging to predict dementia-related outcomes, cognitive tests and structural changes in MRI images. To do this we will use the 180,000 eye images in UK Biobank, one of the world’s largest eye-imaging studies.

    In contrast with the existing literature, we will use deep learning methodologies to improve the ability to extract information from structural eye imaging. We will focus on prediction of a range of cognition outcomes, reflecting real world usage. The UK Biobank also contains brain MRI imaging. We can use this to explore the ability of structural changes in the eye to predict structural changes in the brain.

    Chief Investigator: Benjamin Goudey, School of Computing and Information Systems, Faculty of Engineering and Information Technology

    MDAP Collaboration Leads: Daniel Russo-Batterham & Zaher Joukhadar 

  • Designing and Implementing an Interactive Data Platform for the Monitoring and Evaluating Climate Communication and Education (MECCE) Project

    The MECCE Project is a global collaboration of more than 80 leading researchers and agencies, including UNESCO, the UN Framework Convention on Climate Change, and the International Panel on Climate Change. It aims to increase the quality and quantity of climate communication and education to advance global climate literacy and action.

    This collaboration will support the design and implementation of an open-source Interactive Data Platform (IDP), a key component of the MECCE Project research impact strategy. The IDP will act as a digital interface for datasets and research arising from the MECCE Project. The contributions from MDAP will include IDP scoping, design, development and release, and strategic planning to ensure the IDP has a long-term impact.

    Chief Investigator: Marcia McKenzie, Social Transformations of Education, Melbourne Graduate School of Education

    MDAP Collaboration Lead: Geordie Zhang

  • Improving outcomes for people with complex mental health issues: insights from SANE Australia’s online and phone services

    SANE Australia runs a phone and online counselling service that provides peer support, counselling, support, information, and referrals. The service is used by adults who identify as having a complex mental health issue, complex trauma or high levels of psychological distress, as well as the family or friends who support them. 

    SANE has 12 months of data related to these calls. These data include details of calls as well as some details of call content. We will work with SANE to apply data analysis techniques to understand how people with complex mental health issues interact with SANE, and how SANE might better meet their needs. 

    Chief Investigator: Nicola Reavley, Melbourne School of Population and Global Health, Faculty of Medicine, Dentistry & Health Sciences 

    MDAP Collaboration Lead: Aleks Michalewicz 

  • Fit for Students … Using Canvas data to tailor disciplinary eLearning approaches

    How can discipline-specific pedagogies be translated to eLearning platforms? This collaboration uses topic modelling and sentiment analysis of Canvas data to identify key approaches to teaching and learning practices across multiple disciplines.

    Subject data will be contextualised using University handbook entries and related to student experiences via end of subject survey responses. Built Environments Learning + Teaching (BEL+T) ‘s Delivery, Interaction and Assessment (DIA) model will provide a framework for analysis, and findings will be tested with subject coordinators. Data-based heat maps will offer visual summaries of eLearning activity and inform refinements to teaching practices.

    This project will offer the University and academics an overview of Canvas use in different disciplines. It will also provide a foundation for live feedback tools to inform improvements to online learning practices. 

    Chief Investigator: Kate Tregloan, Built Environment Learning and Teaching, Architecture, Building and Planning

    MDAP Collaboration Leads: Kristal Spreadborough & Amanda Belton 

  • Caring for data now and for the future: Indigenous artistic cultural heritage data management, curation for access, and sustainability

    The Wilin Centre for Indigenous Arts and Cultural Development and Research Unit for Indigenous Arts and Cultures (RUIAC) hold several collections of data, including video, audio, photographs, and text-based documentation.

    In common with many Indigenous cultural, arts, and language centres across the globe, Wilin/RUIAC faces challenges to ensure:

    1. That this cultural heritage data is accessible in an appropriate form to current and future generations of cultural custodians, and to researchers
    2. New knowledge generated by research use and community use is linked to and enriches source records.

    These collections record and describe tangible and intangible cultural heritage created and performed by visiting Indigenous artists and artist-researchers at artistic research events at the Boonwurrung/Southbank campus of UoM.

    This project addresses the complex question: How can rich, living, cultural heritage data be preserved and managed in a university environment in a way that:

    • supports cultural vitality
    • protects IP, ICIP, moral rights, and copyrights of creators and communities
    • supports new community and research use?

    Chief Investigator: Sally Treloyn and Tiriki Onus, Wilin Centre for Indigenous Arts and Cultural Development, Faculty of Fine Arts and Music

    MDAP Collaboration Lead: Emily Fitzgerald 

Previous collaborations

If you would like to view our previous collaborations, please step this way.