Australian BioCommons Webinar series: AI in the Life Sciences:Exploring Possibilities, Inspiring Change

[Giorgia Mori]

Hi everyone,

Australian BioCommons is pleased to invite you to our webinar series, "AI in the Life Sciences: Exploring Possibilities, Inspiring Change."

Building on the foundational webinar led by Dr. Benjamin Goudey in March 2025 (you can watch the recording here), you'll gain insights into real-world applications from leading research experts. Our aim is to inspire fresh perspectives and help you explore the potential of AI in a way that is both accessible and relevant to your needs.

When: 11 June to 2 September 2025

Free and open to anyone associated with an Australian organisation.

More details and registrations

Feel free to pass this onto friends and colleagues. You can also share on LinkedIn!

We hope to see you there!

All my best,

Giorgia

--

Giorgia Mori, PhD (she/her)

BioCloud Training & Communication Officer
Australian BioCommons

The University of Melbourne

M: +61 (0) 466344648

W: biocommons.org.au

My position is funded by Bioplatforms Australia and hosted by The University of Melbourne.

I acknowledge the Traditional Owners of the lands upon which I work and pay my respects to their Elders, both past and present.

I work flexibly so this email may arrive out of hours - please respond at a time that suits you.

[Giorgia Mori]

Hi everyone,

As part of the 'AI for the Life Sciences' webinar series (see email below), I wanted to share details about our next speaker, which may be of particular interest to the single-cell community.

Speaker: Stefano Mangiola, Group Leader, South Australian Immunogenomics Cancer Institute

Title: 'Improving the interpretability of AI models for cell biology and precision medicine'

Abstract:

The Human Cell Atlas (HCA), with its millions of single-cell profiles, combined with advances in AI, has enabled the development of foundational models that capture complex biological principles—from cell states to regulatory dynamics. These models excel at tasks such as cell typing, batch correction, and predicting gene perturbation outcomes, yet their capacity to expose the biological patterns they learn—such as lineage hierarchies, disease-driving interactions, or microenvironmental cues—remains largely unexplored. While existing models achieve high predictive accuracy, their interpretability—the ability to trace predictions to biologically meaningful features like gene modules, or regulatory networks—is often an afterthought. This opacity limits their utility for mechanistic discovery, stifling collaboration between computational and experimental researchers who require actionable, testable insights. Our research group is dedicated to systematically evaluating and improving the interpretability of AI models in single-cell biology.

Time and Date: 12-1 pm AEST, Tuesday, August 26

More details and registrations

Feel free to share on LinkedIn with your colleagues!

I hope to see you there!

All my best,

Giorgia