Non-globular proteins in the era of Machine Learning

COST Action CA21160

ML4NGP WORKSHOP

THIS EVENT HAS PASSED.

Resources and infrastructure to explore the continuum between globular and non-globular proteins

March 18-19, 2025

Barcelona, Spain

DESCRIPTION

The workshop was strategically scheduled just before the start of the Elixir 3DBioinfo 2025 meeting, providing an excellent opportunity to discuss the challenges and opportunities in exploring the continuum between globular and non-globular proteins, including IDPs, tandem repeat proteins, protein aggregation, and more. Our goal was to identify commonalities and foster synergies among key resources and databases such as CATH, TED, PDBeKB, PFAM/InterPro, MobiDB, PED, DisProt, Aggrescan4D and RepeatsDB.

The workshop also served as a platform to discuss examples provided by ML4NGP members in which AlphaFold predicts biologically incorrect models. The short-term goal for this initiative was to gather relevant information that will used to produce a community paper to caution scientists against blindly accepting AlphaFold results without critical evaluation.

Group photo with the participants of the ML4NGP workshop held in Barcelona, Spain, on 18 and19 March 2025.

Participants

  • Gonzalo Parra, Barcelona Supercomputing Center, Spain
  • Alexander Monzon, University of Padova, Italy
  • Silvio Tosatto, University of Padova, Italy
  • Maria Cristina Aspromonte, University of Padova, Italy
  • Christine Orengo,  University College London, UK
  • Nicola  Bordin, University College London, UK
  • Gabor Erdos, Eötvös Loránd University, Hungary
  • Salvador Ventura, Autonomous University of Barcelona, Spain
  • Javier Garcia-Pardo, Autonomous University of Barcelona, Spain
  • Typhaine Paysan-Lafosse, European Bioinformatics Institute, UK
  • Jennifer Fleming, European Bioinformatics Institute, UK
  • Juan Cortes, LAAS-CNRS, France
  • Miguel Andrade, Johannes Gutenberg University, Germany
  • Zarifa Osmanli, University of Padova, Italy
  • Sofia Duarte, Universidad Nacional del Litoral, Argentina
  • Khalil  Joron, Hebrew University of Jerusalem, Israel
  • Pablo Mier, Universidad Pablo de Olavide, Spain
  • San Hadži, University of Ljubljana, Slovenia
  • Isabella Felli, CERM, University of Florence, Italy
  • Oriol Barcenas, Autonomous University of Barcelona, Spain
  • Rafayel Petrosyan, American University of Armenia, Armenia
  • Viktor Bartošík, Masaryk University, Czech Republic
  • Gavin Farrell, University of Padova, Italy
  • Pavel Kadeřávek, Masaryk University, Czech Republic
  • Miguel Romero, Barcelona Supercomputing Center, Spain
  • Tamas Hegedus, Institute of Biophysics and Radiation Biology, Semmelweis University, Hungary
  • Marija Vidović, Institute of Molecular Genetics and Genetic Engineering, Serbia
  • Simone Attanasio, Université libre de Bruxelles, Belgium

PRELIMINARY PROGRAM

MARCH 18, 2025

9:00 – 9:15

Registration & Welcome desk

Sign the COST Action attendance sheet

9:15 – 9:30

Welcome note and goals – Alexander Monzon/Dirk Linke

Session 1: AlphaFold “failing” examples (Chair: Alexander Monzon)

9:30 – 11:00

Present your submitted example in 5 minutes:

What are the biological reasons behind this example of “failure”? Focus on discussing structural/functional/evolutionary aspects.

  • Gabor Erdos
  • Juan Cortes
  • Miguel Andrade
  • Pablo Mier
  • Pavel Kadeřávek / Viktor Bartošík
  • Khalil Joron
  • San Hadži
  • Isabella Felli
  • Salvador Ventura / Oriol Barcenas Lopez
  • Rafayel Petrosyan
  • Tamas Hegedus
  • Marija Vedovic

Questions and Discussion

11:00 – 11:30

Coffee break

11:30 – 13:00

Defining Action Items and responsibilities to write a community paper

13:00 – 15:00

Lunch break

Session 2: DBs synergies – Part I (Chair: Silvio Tosatto)

15:00 – 16:00

Short talks (15 min):

Provide an overview of key features and data of each resource.

  • The Encyclopedia of Domains (TED) – Nicola Bordin
  • CATH database – Christine Orengo
  • RepeatsDB – Zarifa Osmanli
  • InterPro/PFAM – Typhaine Paysan-Lafosse

Q&A

16:00 – 16:30

Coffee break

16:30 – 18:00

Open discussion about synergies

  • Manage of non-globular regions
  • Data sharing
  • Cross Linking
  • Sustainability

MARCH 19, 2025

Session 3: DBs synergies – Part II (Chair: Gonzalo Parra)

9:30 – 10:30

Short talks (15 min):

Provide an overview of key features and data of each resource. 

  • Aggrescan4D – Salvador Ventura
  • MobiDB/Protein Ensemble Database (PED) – Alexander Monzon
  • DisProt database – Maria Cristina Aspromonte
  • PDBe-KB and AlphaFoldDB – Jennifer Fleming

Q&A

10:30 – 11:00

Coffee break

11:00 – 12:00

Final Discussion and Future Activities

Location & venue

VENUE

Barcelona Supercomputing Center (Sala 1-3-2)

Plaça d’Eusebi Güell, 1-3, Les Corts, 08034 Barcelona

ORGANIZATION

Scientific COMMITTEE

Gonzalo Parra, Barcelona Supercomputing Center, Spain

Alexander Monzon, University of Padova, Spain
Silvio Tosatto, University of Padova, Spain

This event is part of the activities of the COST Action ML4NGP, CA21160, supported by COST (European Cooperation in Science and Technology).