Title
Machine Learning good practices – DOME recommendations for better Machine Learning in Computational Biology
Workshop details
- Date: Monday, September 12
- Time: 13:30 to 16:30 CEST
- Format: Virtual
Organisers
- Jennifer Harrow, ELIXIR Hub (United Kingdom)
- Fotis E. Psomopoulos, CERTH (Greece)
- Silvio Tosatto, Università degli Studi di Padova (Italy)
- Leyla Jael García-Castro, ZB MED Information Centre for Life Sciences (Germany)
Topic
This workshop will focus on :
- Standards for reporting Machine Learning approaches on Life Sciences, particularly the DOME recommendations
- FAIRness for Machine Learning
- Best practices around Machine Learning
Abstract
Large amounts of biological data are continuously created, processed and transformed. Machine Learning (ML) algorithms have become one of the preferred approaches for data processing and understanding given their capacity to deal with large amounts of data. In Life Sciences, ML is used in a variety of fields with the potential of resulting in ground-breaking medical applications. However, there are some issues around reproducibility, transparency, explainability and so on that could, at least partially, be alleviated by consistently applying good practices reflected in good descriptions and documentation wrt the ML model. This workshop/SIG deals with good practices, guidelines and metadata for FAIR, reproducible and transparent ML in Life Sciences. In this first edition, we will focus on the Data, Optimization, Model and Evaluation (DOME) recommendations for supervised ML in biology which aim at facilitating the assessment of the quality and reliability of reported models.
Invited speakers
Fotis Psomopoulos
Principal Investigator at the Institute of Applied Biosciences, Centre for Research and Technology Hellas (CERTH).
Daniel S. Katz
Chief Scientist at the National Center for Supercomputing Applications (NCSA), Research Associate Professor in Computer Science (CS), Research Associate Professor in Electrical and Computer Engineering (ECE), Research Associate Professor in the School of Information Sciences (iSchool), and Faculty Affiliate in Computational Science and Engineering (CSE) at the University of Illinois Urbana-Champaign.
Macha Nikolski
Head of the Computation Biology and Bioinformatics Lab at the CNRS institute in Bordeaux.
Target Audience
Researchers and research software engineers working on Machine/Deep Learning approaches
Programme
| TIME | CONTENT |
|---|---|
| 13:30 – 13:40 | Welcoming and introduction to the workshop Leyla Jael Castro |
| 13:40 – 13:55 | Standards for reporting ML in LS Fotis Psomopoulos |
| 13:55 – 14:10 | FAIRness for ML Daniel S. Katz |
| 14:10 – 14:15 | Break |
| 14:15 – 14:30 | Challenges for AI in life sciences Macha Nikolski |
| 14:30 – 14:45 | Generalizability through Heterogeneity Purvesh Khatri |
| 14:45 – 15:00 | Invited talk Chas Nelson |
| 15:00 – 15:05 | Break |
| 15:05 – 15:15 | Overview on challenges and opportunities (input gathered from participants) Leyla Jael Castro |
| 15:15 – 16:15 | Panel discussion on challenges and opportunities Invited speakers and panelists |
| 16:15 – 16:20 | Break |
| 16:20 – 16:30 | Wrap-up Leyla Jael Castro |
| 16:30 | Coffee break and finding the next workshop location |