Non-globular proteins in the era of Machine Learning

COST Action CA21160

1ST ML4NGP TRAINING SCHOOL

THIS EVENT HAS PASSED.

Protein aggregation, intrinsic disorder and phase separation in the era of machine learning

April 19-21 2023

Porto, Portugal

Royalty free images.

DESCRIPTION

The COST Action ML4NGP is pleased to announce its 1st Training School on “Protein aggregation, intrinsic disorder and phase separation in the era of machine learning”. The event was co-organized with the H2020-funded Twinning project PhasAGE (GA 952334) and the H2020-MSCA-RISE project IDPFun (GA 778247), and took place at the i3S, in Porto (Portugal), from 19 to 21 April 2023.

The 1st ML4NGP Training School was organized in parallel with the PhasAGE Training School and IDPfun Training School with shared theoretical lectures and computational practical sessions.

The main goal of this training school was to provide participants with an overview of recent advances in computational resources and hands-on training with resources and tools for studying protein aggregation and phase separation. Participants were challenged to reflect on how recent advances in protein structure ML prediction tools are paving the way for a more comprehensive understanding of protein function.

The scientific program included lectures and practical sessions. A dinner and social event was also organized for participants.

The program also included a career development workshop on alternative career paths in biophysics and computational structural biology to help early-stage researchers maximize their professional potential.

Group photo with the participants of the first Training School held in Porto, Portugal from 19 to 21 April 2023.

TARGET AUDIENCE AND REQUIREMENTS

Exclusive for working group members. Participants were selected by the organizing committee in order to guarantee gender balance, early career researchers (PhD students and Junior postdocs) and geographic distribution. The Training School was limited to a reduced number of attendees to maximize active learning. 

This Training school was intended for those who are interested in learning how to better predict and correlate protein structure and function. Participants were expected to navigate online biological resources but were not expected to have computational or programming skills. Prior knowledge and basic training in Biochemistry, Biology, Bioengineering or related sciences were recommended. 

FINAL PROGRAM

19 APRIL 2023

09:00-10:00

AI revolution and protein structure models
Pedro Beltrão

10:00-11:15

PhasAGE-ML4NGP-IDPfun Computational Session – part I

11:15-11:30

Coffee Break

11:30-12:45

PhasAGE-ML4NGP-IDPfun Computational Session – part II

12:45-14:00

Lunch

14:00-15:30

Mathematical models and data analysis
Pedro Martins

15:30-16:00

Modulators of Ataxin-3 protein aggregation

Sandra Ribeiro

16:00-16:15

Coffee Break

16:15-17:15

Career Development Talk
Peter Tompa

17.30-22:00

Social event and Dinner

20 APRIL 2023

10:00-11:00

The structure-toxicity relationship of protein aggregates
Fabrizio Chiti

11:00-11:15

Coffee Break

11:15-12:15

Conformational ensembles of IDPs: implications for function and drug development
Peter Tompa

12:15-13:00

iPLUS: on the function of a non-coding sequence as an asset for biopharmaceutics

Alexandra Moreira

13:00-14:30

Lunch

14:30-16:00

ML4NGP-IDPfun Practical Session
Introduction to Machine Learning Methods in Biology
Jovana Kovačević

16:00-16:15

Coffee Break

16:15-17:45

ML4NGP-IDPfun Practical Session
Local Energetic Frustration, Protein Function and Dynamics
Gonzalo Parra

21 APRIL 2023

9:30-10:30

NMR as a tool to study disordered proteins at atomic resolution
Lukas Zidek

10:30-10:45

Coffee Break

10:45-11:45

CAID 2: lessons from the second critical assessment of protein intrinsic disorder prediction
Silvio Tosatto

12:00-13:00

Disordered interactions and phase separations of α-synuclein
Alfonso de Simone

13:00-14:30

Lunch

14:30-16:00

Trainees presentations  & closing remarks

Practical Sessions

Introduction to Machine Learning Methods in Biology | Jovana Kovačević

In this practical session, participants will be introduced to the basic concepts and techniques of machine learning and their applications in biology with a special emphasis to non-globular proteins. The session will begin with an overview of machine learning and will focus mainly on supervised learning methods. Participants will then learn how to use popular machine learning tools and libraries, such as Pandas, to analyze biological datasets. They will work through several hands-on examples, including protein disorder prediction and disease classification using genomic data.

Local Energetic Frustration, Protein Function and Dynamics | Gonzalo Parra

In this practical session, participants will be introduced to the concept of local energetic frustration, derived from the energy landscapes theory for protein folding. The participants will get familiar with how to use local energetic frustration to understand energetic constraints in protein structures both at single instances and in the context of protein families. We will discuss the algorithms to calculate frustration as well as how to interpret results in terms of protein stability and function. The application of these concepts to IDPs and repeat proteins will be discussed.

ORGANIZATION

SCIENTIFIC organizing COMMITTEE

Rita Vilaça (IBMC/i3S, University of Porto, Portugal)
Mónica Marques (IBMC/i3S, University of Porto, Portugal)
Sandra Ribeiro (IBMC/i3S, University of Porto, Portugal)
Alexander Monzon (University of Padova, Italy)
Silvio Tosatto (University of Padova, Italy)

CONSORTIA

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

FEEDBACK FROM PARTICIPANTS

I enjoyed very much listening to the presenters. They covered interesting topics, most of them related to my study area.

Testimonial #1 1st ML4NGP Training School

I found the lecture on local energetic frustration particularly interesting. Also Peter Tompa really got me thinking a lot. Of course, I would also highlight socializing and interesting talks with other participants.

Testimonial #2 1st ML4NGP Training School

All the talks and practical sessions were of the highest level.

Testimonial #3 1st ML4NGP Training School