Poster Presentation 29th Annual Lorne Proteomics Symposium 2024

Analysis platforms for Australian proteomics (#112)

Johan Gustafsson 1 2 , Matt P Padula 3 , Cameron Hyde 4 , W C Mike Thang 4 , Ben Crossett 5 , Pawel Sadowski 6 , Matthew Downton 7 , Sarah Beecroft 8 , Mark Gray 8 , Ziad Al Bkhetan 1 2 , Gareth Price 4 , Nigel Ward 2 4 , Steven Manos 1 2 , Cali E Willet 9
  1. Melbourne Bioinformatics, University of Melbourne, Melbourne, Victoria, Australia
  2. Australian BioCommons, https://www.biocommons.org.au/
  3. School of Life Sciences, University of Technology Sydney, Sydney, New South Wales, Australia
  4. QCIF Facility for Advanced Bioinformatics, Institute of Molecular Biology, University of Queensland, Brisbane, Queensland, Australia
  5. Sydney Mass Spectrometry, The University of Sydney, Sydney, New South Wales, Australia
  6. Central Analytical Research Facility, Queensland University of Technology, Brisbane, Queensland, Australia
  7. National Computational Infrastructure (NCI) , Canberra, Australian Capital Territory, Australia
  8. Pawsey Supercomputing Research Centre, Perth, Western Australia, Australia
  9. Sydney Informatics Hub, Core Research Facilities, University of Sydney, Sydney, New South Wales, Australia

The transition of proteomics into a data rich discipline now demands access to scaled up compute and complex analysis pipelines. There is an opportunity to respond as a coordinated community to provide fit for purpose analysis platforms that will support Australian proteomics moving forward.

Today, the application of bioinformatics to proteomics is complicated by numerous and diverse software tools, the steep learning curve in the adoption of working on the command line, limited access to computational resources, and intimidating pathways to larger scale compute resources (e.g. national computational systems). Addressing these challenges is critical given the increase in both experimental complexity and data output size in proteomics. Solutions to these challenges include supporting access to fit-for-purpose software and compute systems, the details of which are defined by the proteomics community.

Since the finalisation of the Proteomics Bioinformatics Infrastructure Roadmap for Australia in 2022 [1], the Australian BioCommons has been leading the effort to tailor software and compute resources for proteomics, in close collaboration with the community and national compute providers. To date, the two key outcomes of this effort are the Proteomics Lab on Galaxy Australia [2] and the Australian BioCommons Leadership Share (ABLeS) [3].

  1. The Galaxy Australia Proteomics Lab provides researchers with a fully-subsidised shared platform for point-and-click access to key software (MaxQuant, DIA-NN), interactive visualisation tools (LFQanalyst) and reference data that is specific to proteomics. Galaxy Australia is underpinned by national computational resources, and will be expanded where possible to accommodate day-to-day analysis requirements for proteomics researchers.
  2. ABLeS is a novel access mechanism for national computing facilities (Pawsey Supercomputing Centre and National Computational Infrastructure (NCI)), supporting bioinformaticians to access compute at scale for software optimisation, production implementation of standard methods and creation of critical reference data. ABLeS intends to accelerate scaling and application of bioinformatics on national facilities, and to share the learnings from this work with the wider life science community. ABLeS is currently being used for large-scale testing and optimisation of DIA-NN at the NCI.

This presentation will outline how ABLeS and Proteomics Lab work, and how the Australian BioCommons and its partners will continue to evolve these activities in lock step with the changing requirements of Australian proteomics research.