Aperçu des sections
-
Overview
Artificial intelligence (AI) has permeated our lives, transforming how we live and work. Over the past few years, a rapid and disruptive acceleration of progress in AI has occurred, driven by significant advances in widespread data availability, computing power and machine learning. Remarkable strides were made in particular in the development of foundation models - AI models trained on extensive volumes of unlabelled data. Moreover, given the large amounts of omics data that are being generated and made accessible to researchers due to the drop in the cost of high-throughput technologies, analysing these complex high-volume data is not trivial, and the use of classical statistics can not explore their full potential. As such, Machine Learning (ML) and Artificial Intelligence (AI) have been recognized as key opportunity areas for ELIXIR, as evidenced by a number of ongoing activities and efforts throughout the community.
However, beyond the technological advances, it is equally important that the individual researchers acquire the necessary knowledge and skills to fully take advantage of Machine Learning. Being aware of the challenges, opportunities and constraints that ML applications entail, is a critical aspect in ensuring high quality research in life sciences
Recognizing this need, this week-long training will bring together experts from four ELIXIR Nodes and deliver a hands-on, high-intensity course, available for members from all ELIXIR Nodes.
Learners will be guided across the various steps in Machine Learning, from the foundational concepts, through the deep learning and generative AI techniques, closely complemented by insights into the existing reporting (DOME Recommendations) and regulatory frameworks (EU AI Act).
This 4-day school will involve around ten trainers/helpers from across 4 different ELIXIR nodes and 30 participants from across all ELIXIR nodes. It will be hosted in France in May 2025 (see below for venue and time).
The registration fees (including accommodation in single rooms and meals) are 650 EUR for academics and 1000 EUR for for-profit companies.Audience
This course is addressed to bioinformaticians, biostatisticians, bioanalysts, life scientists and biomedical researchers with good python programming skills and general knowledge of Machine Learning approches.
Datasets
Applications will be agnostic of data types and species. The hands-on datasets used in the course will be aligned with participants’ preferences wherever feasible.
Learning outcomes
At the end of the course, the participants should be able to:
- understand AI approaches
- be aware of regulations and standards of AI
- start applying learned approaches to their own data
Prerequisites
Knowledge / competencies
The level of this course is intermediate, with the following requirements:- experience with data analysis
- Intermediate Python programming
- Machine Learning basics
Technical
You are required to bring your own laptop, more instructions will be communicated to the course participants.Preliminary schedule
(detailed agenda to be circulated closer to the date)
D1 - Monday 19 May
Afternoon: Introduction to Machine Learning
D2 - Tuesday 20 May
Morning: Introduction to Neural Networks
Afternoon: More Neural Networks
D3 - Wednesday 21 May
Morning: Introduction to Deep Learning
Afternoon: More Deep Learning
D4 - Thursday 22 May
Morning: Introduction to Generative AI and Large Language Models
Afternoon: Regulation/standards for AI
D5 - Friday 23 May
Morning: Bring Your Own DataApplications
Applications are now open at https://framaforms.org/ai-ml-in-life-sciences-application-form-1734346608. The number of participants will be limited to 30. Applications will be evaluated based on prerequisite skills and knowledge, as well as motivation for the training.
We will need to start the validation process late January. Thank you for your understanding. Here is provisional agenda:
December 18, 2024: Call for Application opens
January 31, 2025: Call for Application closes
February 21, 2025: Successful applicants announced
May 19, 2025: School starts
Venue and Time
This course will take place in the Centre de Vacances et Colloques Paul Langevin (CNRS), in the village of Aussois (Savoie, France).
The event will start on Monday May 19th 2025 after lunch and end on Friday May 23rd 2025 around lunch time.
Additional information
Coordination: Lucie Khamvongsa-Charbonnier (IFB, ELIXIR-FR), Olivier Sand (IFB, ELIXIR-FR)
Scientific committee: Bérénice Batut (IFB, ELIXIR-FR), Alexander Botzki (VIB, ELIXIR-BE), Robbin Bouwmeester (VIB, ELIXIR-BE), Wandrille Duchemin (SIB, ELIXIR-CH), Styliani-Christina Fragkouli (CERTH, ELIXIR-GR), Ralf Gabriels (VIB, ELIXIR-BE), Raphaël Mourad (University of Toulouse III, ELIXIR-FR), Fotis Psomopoulos (CERTH, ELIXIR-GR), Harikrishnan Ramadasan (VIB, ELIXIR-BE), Thuong Van Du Tran (SIB, ELIXIR-CH)
IFB abides by the ELIXIR Code of Conduct. Participants of IFB courses are also required to abide by the same code. Please, make sure that you read it before the event.
For more information, please contact
contact-formation@groupes.france-bioinformatique.fr - understand AI approaches
-
Hours Title Main instructor/presenter Support Day 1 14:00 - 14:10 Welcome Olivier Sand Slides 14:10 - 14:20 Icebreaking activity Bérénice Batut Cards 14:20 - 14:50 Participant presentations Slides 14:50 - 18:50 Intro to ML Wandrille Duchemin 14:50 - 15:50 General sklearn syntax 15:50 - 16:10 Coffee break 16:10 - 16:55 Overfit/underfit, the need for regularization,
cross validation and a test set16:55 - 17:40 Hands-on: overfit/underfit, the need for regularization,
cross validation and a test setTutorial 17:40 - 17:50 Short break 17:50 - 18:20 Metrics and imbalance 18:20 - 18:50 Hands-on: Metrics and imbalance 19:30 Dinner Day 2 09:00 - 09:15 Recap from day 1 Wandrille Duchemin 09:15 - 09:30 ELIXIR Fotis Psomopoulos 09:30 - 12:30 Intro to Neural networks Ralf Gabriels & Hari Ramadasan 09:30 - 10:30 Logistic regression (in pytorch)
on a classification task10:30 - 11:00 Coffee break 11:00 - 12:30 Neural networks (in pytorch) on a
classification task12:30 - 14:00 Lunch 14:00 - 17:30 Intro to DL (without GAI) 14:00 - 16:00 CNN (in pytorch) on a classification 16:00 - 16:30 Coffee break 16:30 - 17:30 RNN with attention (in pytorch)
on a regression task19h30 Dinner Day 3 09:00 - 09:15 Recap from day 2 Ralf Gabriels & Hari Ramadasan 09:15 - 12:30 Intro to GAI and LLM Raphaël Mourad 09:15 - 10:30 Introduction to LLM 10:30 - 11:00 Coffee break 11:00 - 11:45 Pretraining LLM for DNA - hands on Tutorial 11:45 - 12:30 Hands-on: Finetuning LLM,
zeroshot prediction for DNA variants,
synthetic DNA sequence generationTutorial
Tutorial
Tutorial12:30 - 14:00 Lunch 14:00 - 17:30 Regulations/standards for AI - DOME Fotis Psomopoulos Tutorial 14:00 - 14:30 Introduction to common challenges
in current AI/ML applications14:30 - 15:00 Hands-on: group activity;
review selected articles
and assess the (potential) issues - if any15:00 - 15:15 Introduction to the DOME
recommendations15:15 - 16:00 Hands-on:
annotation of the previously selected articles
- assessment of the FAIR aspects - part 116:00 - 16:30 Coffee break 16:30 - 17:00 Hands-on:
annotation of the previously selected articles
- assessment of the FAIR aspects - part 217:00 - 17:30 Review of the key points of the AI Act
and the GenAI guidelines18:45 Savoyard Apéritif 19:30 Savoyard Dinner Day 4 09:30 - 12:30 General recap / Supplementary exercises All 09:30 - 10:20 Extra ML 10:20 - 10:50 Coffee break 10:50 - 11:40 Neural networks (in pytorch lightning) 11:40 - 12:30 Extra LLM 12h30 - 14:00 Lunch 14:00 - 17:00 Bring Your Own Question/Data/Model All 14:00 - 16:00 Group work 16:00 - 16:30 Coffee break 16:30 - 18:30 Group picture / Walk or work? 19:30 Dinner Day 5 09:00 - 10:30 Group work 10:30 - 11:00 Coffee break 11:00 - 12:00 Group work / presentations 12:00 Lunch