The F2F AI in Omics Workshop will be held at the University of Bradford. Organised by members affiliated with the UKRI Digital Skills Catalyst: Pasky Miranda (University of York) and Khaled Jum'ah (University of Bradford).

Programme Agenda

Event Details
This one-day event (organised by UKRI Digital Skills Catalyst) will support AI readiness for bioscientists.

The morning includes two talks:

Lecture: Introduction to Machine Learning Concepts Used in -Omics
Description:
This session will introduce participants to the core principles of supervised machine learning (ML) as applied to omics data analysis. The lecture will cover the unique challenges posed by high-dimensional and complex omics datasets, including issues of reproducibility, overfitting, and the pitfalls of poor feature selection. Participants will learn about best practices for robust and interpretable ML workflows and will be introduced to reporting standards such as DOME and FAIR. The session will include a practical critical appraisal of published examples, to help attendees identify good and bad practice in real-world contexts.

Practical Demo: Pitfalls in ML and Good Practice Using Tidymodels
Description:
Building on the introduction, this live session demonstrates common pitfalls in ML workflows for omics and shows how the Tidymodels framework in R promotes best practice. Through a guided tutorial, participants will learn how to build a robust ML pipeline: from preprocessing and avoiding data leakage, to model fitting and proper evaluation. Participants will follow along with a Quarto notebook, applying what they learn to a classification and regression problems relevant to omics.
The afternoon will consist of a bring your own data Code Retreat where our team of experts will be on hand to help you with your personal projects, and a Hackathon challenge activity for those without their own data who want to try out their new skills.

Networking tea/coffee and lunch will be included.