GFM for Diagnostic Innovation 

Time: 15 December 2025, 08:50–12:30 (Wuhan local time)

Format: Hybrid (In-person + Microsoft Teams)

In-person venue: Grand Ballroom 1, Wanda Realm Wuhan – 3F, Wanda Reign Wuhan

Online link: BIBM Workshop on Genomic Foundation Models for Diagnostic Innovation | Meeting-Join | Microsoft Teams.

Workshop Description

Overview

The landscape of precision medicine and computational biology is undergoing a rapid transformation with the advent of Genomic Foundation Models (GFMs) — powerful AI systems inspired by large-scale models from natural language processing and computer vision. GFMs are designed to learn generalisable, high-resolution representations of genomic data, enabling intelligent and scalable solutions to some of the most pressing challenges in laboratory diagnostics and clinical decision-making.

About the Workshop

This international workshop, held as part of the IEEE BIBM 2025 Conference, is organised under the framework of the EU-funded GenDAI project (Genomic Applications for Laboratory Diagnostics supported by Artificial Intelligence). It aims to bring together a multidisciplinary community of researchers, clinicians, AI specialists, and industry stakeholders to advance the field of Genomic Foundation Models.

Participants will have the opportunity to engage with cutting-edge research, share practical insights, and identify future directions for integrating foundation models in clinical genomics. By facilitating meaningful dialogue between AI and bioinformatics communities, the workshop seeks to accelerate the translation of research breakthroughs into deployable, ethical, and trustworthy solutions in real-world healthcare contexts.

Key Themes and Topics

The workshop will feature presentations, panel discussions, and paper sessions covering (but not limited to) the following areas:

• Design and pretraining of Genomic Foundation Models: Model architectures, tokenisation of genomic sequences, pretraining corpora, and scalability considerations
  
  

• Fine-tuning strategies for diagnostic applications: Tailoring GFMs for disease classification, variant prioritisation, and patient stratification
  
  

• Multi-omics integration using transformer-based architectures: Combining genomic, transcriptomic, epigenomic, and clinical data for holistic insights
  
  

• Deployment of GFMs in laboratory and clinical environments: Case studies, workflow integration, and automation
  
  

• Evaluation frameworks and benchmarking practices: Standardisation, reproducibility, and performance metrics
  
  

• Federated learning and privacy-preserving model development: Ensuring secure and decentralised learning with sensitive biomedical data
  
  

• Ethical, legal, and societal implications of GFMs: Transparency, fairness, and alignment with regulatory frameworks
  
  

• Bias detection and mitigation strategies: Addressing data imbalance and equity in genomic datasets
  
  

• Case studies from the GenDAI consortium and beyond
  
  

Who Should Attend?

This workshop welcomes:

• AI and machine learning researchers working on biomedical applications
  
  

• Bioinformaticians and computational biologists
  
  

• Clinical genomics and laboratory medicine professionals
  
  

• Data scientists from healthcare, pharma, and biotech industries
  
  

• Regulators and policy experts interested in ethical AI in genomics
  
  

• Graduate students and early-career researchers seeking exposure to state-of-the-art research
  
  

Format and Participation

The workshop will be held in a hybrid format — onsite in Wuhan, China, with full virtual participation options available. All sessions will be streamed live, with opportunities for remote presentations and interactive Q&A. Recordings will also be made available to registered participants post-event.

The organising team will ensure a seamless experience for all attendees, regardless of location, and encourage broad international engagement.

Important Dates

• 25 June 2025: Call for Workshop Papers issued

• 15 October 2025 (extended to 31 October 2025): Full paper submission deadline

• 10 November 2025: Notification of acceptance to authors

• 23 November 2025: Final camera-ready papers due

• 15–18 December 2025: Workshop held at IEEE BIBM 2025, Wuhan, China

Paper Submission Instructions

Submission URL: https://wi-lab.com/cyberchair/2025/bibm25/index.php 

or directly here: https://wi-lab.com/cyberchair/2025/bibm25/scripts/submit.php?subarea=S32&undisplay_detail=1&wh=/cyberchair/2025/bibm25/scripts/ws_submit.php

Types of Papers: We invite long and short research papers. Long paper submissions are expected to describe substantial, original, completed, and unpublished work. Characteristics of short papers include a small, focused contribution; a work in progress; a negative result; an opinion piece; or an interesting application nugget. 

Paper Length: As per  IEEE BIBM 2025 policy, Long research papers may consist of up to 8 pages of content, including references and appendices; final versions of long papers will be given one additional page of content (up to 9 pages) so that reviewers’ comments can be taken into account. Short research papers may consist of up to 4 pages of content; final versions of short papers will be given one additional page of content (up to 5 pages) so that reviewers’ comments can be considered.

Submission Formatting: Please follow the IEEE BIBM 2025 official LaTeX or Word templates 

Please submit a full-length paper (up to 8 page IEEE 2-column format) through the online submission system (you can download the format instruction here) http://www.ieee.org/conferences_events/conferences/publishing/templates.html. 

The paper submission is double blind. Please don’t include any authors and/or affiliations in the paper.

Blind Reviewing Policy: The workshop follows a blind reviewing policy. The authors should omit their names and affiliations from the paper and avoid self-references that reveal their identities. Papers that do not conform to these requirements will be rejected without review. Accepted papers will be published in the workshop proceedings.

Double submission policy: Parallel submission to other meetings or publications is possible but must be immediately notified to the workshop organizers.

For further information, please contact Haithem Afli (Haithem.Afli@mtu.ie)

Workshop Organizers                                       

Programme Co-Chairs : 

• Dr Haithem Afli, ADAPT Centre, Munster Technological University, Ireland

Programme Committee Members : 

• Professor Matthias Hemmje, FTK Research Institute, Germany

• Dr Bruno Andrade, ADAPT Centre, Munster Technological University, Ireland

• Dr Thomas Krause, INCONTEC GmbH, Germany

• Mike Leske, ADAPT Centre, Munster Technological University, Ireland

• Professor Haiying Wang, Ulster University, United Kingdom

• Dr Francesca Bottacini, ADAPT Centre, Munster Technological University, Ireland

• Professor Huiru (Jane) Zheng, Ulster University, United Kingdom

Workshop Program 

Opening Session

08:50–09:00
Opening Remarks
Chair: Haithem Afli

09:00–09:20
Keynote: Genomic Foundation Models — From Research to Real-World Diagnostic Innovation
Speaker: Dr Bruno Andrade
Chair: Haithem Afli

Session 1 — Foundations and Infrastructures

Session Chair: Dr Thomas Krause

09:20–09:40
Paper: Evo2-Virus: Ultra-short Viral Sequence Identification with Evo2
Authors: Zhixiang Xu, Hankai Yang, Xiaoya Fan, Zhong Wang

09:40–10:00
Paper: The GenDAI Cloud-Native Infrastructure and Data Stewardship for Clinical Metagenomic Diagnostics
Authors: Philippe Tamla, Thomas Krause, Matthias Hemmje, Flavia Monti, Francesca De Luzi, Massimo Mecella, Bruno Andrade, Paolo Buono, Andrea Molinari

10:00–10:30
Coffee Break

Session 2 — AI, Ethics and System Design

Session Chair: Philippe Tamla

10:30–10:50
Paper: LLM-driven Cloud-based Infrastructure Design
Authors: Andrea Sepielli, Marco Calamo, Filippo Bianchini, Francesca De Luzi, Matteo Marinacci, Flavia Monti, Jacopo Rossi, Massimo Mecella, Philippe Tamla, Thomas Krause, Matthias Hemmje

10:50–11:10
Paper: Responsible Use of AI in Genomics and Ethical Implications
Authors: Giulia Ricci, Paolo Buono, Thomas Krause, Philippe Tamla, Matthias Hemmje, Francesca De Luzi, Francesco Leotta, Andrea Marrella, Flavia Monti, Massimo Mecella

11:10–11:30
Paper: From Reads to Reports: A Vision for a GFM-Powered Genomic Diagnostic Platform
Authors: Thomas Krause, Philippe Tamla, Andrea Leoni, Flavia Monti, Francesca De Luzi, Jamie FitzGerald, Bruno Andrade, Haithem Afli, Massimo Mecella, Paolo Buono, Andrea Molinari, Matthias Hemmje

Session 3 — Biomarkers, Taxonomy and Clinical Reporting

Session Chair: Dr Bruno Andrade

11:30–11:50
Paper: Genomic Foundation Models for SNP Analysis
Authors: Johanna Pethke, Thomas Krause, Bruno Andrade, Michael Kramer, Matthias Hemmje

11:50–12:10
Paper: Genomic Language Models Applied for Microbial Taxonomy Classification
Authors: Mike Leske, Jamie Fitzgerald, Keith Coughlan, Thomas Krause, Matthias Hemmje, Francesca Bottacini, Haithem Afli, Bruno Andrade

12:10–12:25
Paper: Revisiting Data Visualizations in Diagnostic Reports
Authors: Paolo Buono, Philippe Tamla, Thomas Krause, Francesca De Luzi, Flavia Monti, Massimo Mecella

12:25–12:30
Closing Remarks
Chair: Haithem Afli