Artificial Intelligence and Big Data Approaches in Molecular Medicine Workshop

FULLY BOOKED - If you're curious about how AI and big data will change healthcare-related discoveries, from finding new medicines to understanding our genes, this is the place to be! This event is joined by a symposium on Dynamics and modelling of biological systems by DANEMO.

Info about event

Time

Wednesday 24 January 2024, at 11:30 - Thursday 25 January 2024, at 11:00

Location

Copenhagen University, Denmark

Organizer

Nordic EMBL Partnership

“Artificial Intelligence and Big Data Approaches in Molecular Medicine” workshop by the Nordic EMBL Partnership for Molecular Medicine

Fully booked, registrations are not accepted from 19 December 2023.

Scientific programme
Wednesday, 24 January 2024

Venue: Lundbeckfond Auditorium and Room 4-0-24, Copenhagen Biocenter, Copenhagen, Denmark
 

11:30-12:15 Registration and lunch

12:15-12:30 Welcome and introducing the Nordic EMBL Partnership
Poul Nissen,
 Former Director of DANDRITE, Vice-Dean for Research, Innovation & Business Deveopment at Aarhus University, Faculty of Natural Sciences; and Nóra Lehotai, Communications coordinator of the Nordic EMBL Partnership

Moderator: Nóra Lehotai
12:30-13:15 Johan Trygg, Sartorius and Umeå University
Title: AI/Machine learning and Statistics to analyse, interpret and create predictive models in biology and medicine - pitfalls and opportunities

13:15-13:25 Moving to Room 4-0-24 for hands-on part 1

13:25-15:30 Hands-on - part 1, led by Johan Henriksson, MIMS, Umeå University
Topic: Bayesian statistics for handling large complex data

15:30-16:00 Return to auditorium and coffee break

16:00-16:45 Birgit Kriener, NCMM, University of Oslo
Title: Inference of clinically actionable alterations from digital tissue images with deep-learning

16:45-17:30 Nina Linder, FIMM, University of Helsinki
Title: AI for point care diagnostics

17:30-17:40 Concluding the day

 

Thursday, 25 January 2024

Venue: Room 4-0-24, Copenhagen Biocenter, Copenhagen, Denmark

9:00-10:50 Welcome back and hands-on - part 2, led by Johan Henriksson, MIMS, Umeå University

Topic: Neural networks and the link to modern statistics

10:50-11:00 Closing of event

---

11:00-12:00 Lundbeckfond Auditorium, Registration for the “DANEMO Symposium 2024 – Dynamics and modelling of biological system” with coffee, lunch and poster hang up

12:00 Opening of “DANEMO Symposium 2024 – Dynamics and modelling of biological system” - See programme here

Speaker: Professor Johan Trygg

Head of Advanced Data Analytics at Sartorius, Professor at the Department of Chemistry, Umeå University. 

Speaker: Dr Birgit Kriener

Researcher at NCMM, University of Oslo, research focus on computational oncology. 

Speaker: MD, PhD Nina Linder

Physician by training. Her research involves the development of novel artificial intelligence-based solutions for cancer and infectious disease diagnostics. Nina is also co-heading several projects developing artificial intelligence-based tools for point-of-care diagnostics in a global setting, such as United Nations. 

Hands-on sessions: Dr Johan Henriksson

Researcher in T-cell biology and development of new methods (bioinformatics algorithms and wet lab protocols), with special emphasis on immunotherapy and CAR T cells.

Details on the hands-on workshop

All hands-on practice will be done using web-based simulators, to ensure that even those without programming experience can follow. The focus will be on the central concepts and getting a better intuition of all these methods. For those that know R we will also provide some code, so you can then put this into practice afterwards!

Day 1, 14:30-16:30 Hands-on - part 1, led by Johan Henriksson, MIMS, Umeå University
Topic: Bayesian statistics for handling large complex data

Machine learning is effectively Bayesian statistics in disguise. In the first workshop we will look at penalized linear models as an example, exploring the link between priors and hyperparameters. Basic concepts such as L1 and L2 penalization will be introduced, and we will explore their pros and cons. How is machine learning different from normal statistics?

Day 2, 9:00-10:50 Welcome back and hands-on - part 2, led by Johan Henriksson, MIMS, Umeå University

Topic: Neural networks and the link to modern statistics

We will here continue with nonlinear models, and why you want (or don't want!) to use these over linear models. What type of models exist? How do you compute and interpret a ROC curve? What is a hyperparameter? We will touch upon explainable ML, and we will especially dive into what is a good and a bad explanation, and how you can prepare yourself for what you might get out of the models.