Introducing opendata.fit: a FAIR data analysis and publication platform

Mr James Wilmot1,2, Ms Varvara Efremova1,2, Professor Pall Thordarson1,2

1UNSW Sydney, , Australia, 2Australian Characterisation Commons at Scale, , Australia

This talk introduces the opendata.fit platform, developed as part of the Australian Characterisation Commons at Scale (ACCS) project. opendata.fit is a web-based data analysis and publication platform, focused on improving reproducibility in research and supporting FAIR (Findable, Accessible, Interoperable and Reusable) data principles. opendata.fit builds on the work of Bindfit, a web-based binding constant fitting platform developed at UNSW.

opendata.fit aims to provide users with a discipline-agnostic, common platform to manage scientific analysis workflows and datasets from development to publication. It provides tooling to upload instrument data, execute analysis algorithms in the cloud, visualise results, and publish complete workflows as citable entities. Users will be able to choose from a library of existing public workflows or develop their own. It currently supports execution of Python-based analysis algorithms with plans to extend support to other common scientific computing languages.

As part of this work, we have developed a novel approach to data portability and metadata management using and extending the open source Frictionless Data standard and Vega visualisation language.

This presentation will provide an overview of the opendata.fit platform, demonstrate a sample analysis workflow on small angle scattering data, and discuss the platform’s future direction.


Biography:

Varvara is a research software engineer at the University of New South Wales with a background in astrophysics. She has extensive experience with the Python scientific stack, containerisation, web development, as well as cloud computing infrastructure in the context of scientific applications. She is interested in using modern web and containerisation technologies to develop tools that enable consistent, reproducible execution and publication of scientific workflows.

James is a research software engineer at UNSW Sydney with a background in physics. He has extensive experience in web application technologies, containerisation, data structures and metadata. When not coding he enjoys running, cycling and being outdoors.

Both Varvara and James help run and maintain a community makerspace in Canberra.

Categories