Brainhack Marseille

December 6 - December 8 2021


To register to the event, please click on the button below to fill the form. The deadline for registration is Sunday 28 November.
Registration is now closed.
For late registration, you should anyway fill the form (press on the grey 'REGISTER' button) and send an email to the organization team.
Lunch and social event are not guaranteed for late registrations.


Please find bellow the program of the Brainhack event (timezone Paris UTC+1). If you want to add the following schedule to your own personal Google agenda, click on the "Save the dates" button and then click on the "+ Google agenda" located at the bottom right of your screen.

6th December

09h00-10h15 Introduction to BrainHack Marseille 2021
  • 09h00-09h15 - Introduction to the event
  • 09h15-10h15 - Projects pitches
10h15-12h30 Participants can chose to attend one of the two parallel sessions:
Parallel session 1: Hacking
Parallel session 2: Python for non-programmers, a tutorial by Julia Sprenger
12h30-14h00 Break
14h00-16h30 Hacking
16h30-17h30 Unconference - Remi Gau, Marie-Hélène Bourget, Martin Szinte & Sylvain Takerkart

BIDS and its recent extensions

7th December

09h00-11h00 Hacking
11h00-12h30 Marseille-Lyon exchange session: software demos and discussion
12h30-14h00 Break
14h00-17h00 Hacking
17h00-20h00 BHM social event!

8th December

09h00-11h00 Hacking
11h0-12h00 Unconference - Paola Mengotti

Diversity in Academia: The Women in Neuroscience Repository (WiNRepo)

12h00-14h00 Break
14h00-15h00 Hacking
15h00-16h00 Unconference - Greg Operto

Management and Quality Control of Large Neuroimaging Datasets: Developments From the Barcelonaβeta Brain Research Center, (Huguet et al. Frontiers in Neuroscience 2021)

16h00-17h00 Wrapping up about the projects with everybody


Here you can find all the informations about the event projects.
If you want to submit a project you should follow the link, fill the form, and open a github issue. Projects can be anything you'd like to work on during the event with other people (coding, discussing a procedure with coworkers, brainstorming about a new idea), as long as you're ready to minimally organize this!

EyetrackPrep: an automatic tool to pre-process eye-tracking data (fixation, blinks, saccades, micro-saccade, pursuit, pupillometry)

by Martin Szinte, Anna Montagnini & Guillaume Masson

This project aims at providing to the eye-tracking research community a standardized software allowing the automatic computation and quality check of eye-tracking data. This includes any records of eye position metrics such as fixation, eye movements (saccades, microsaccades, pursuit), blinks, and pupillometry.
EyeTrackPrep is meant in the future to be used by all researchers using eye-tracking data. It aims at simplifying and standardizing this domain to improve research reproducibility as well as the share of collected data.
Eye-tracking community lack such a tool even if a need for it is clear. Although this project might take time until a functional software is produced, it will modify for long term the way cognitive neuroscientists work and generate open science of high quality data akin what happened with the development of BIDS-apps from the neuroimaging community.

The following points are going to be addressed in this project:
  • Discussions about the tool, its goals, what it should and what it should not include
  • Modeling of eye movement metrics (e.g. saccade detection from eye-traking position time series)
  • Visualization for data quality check

Required skills

This is a python project, a mix of coding and non-coding skills are required:
Knowledge/interest in eyetracking 100%
Python programming 70%

Integration of the single-trial time-resolved spectral connectivity (coherence; PLV) in Frites

by Vinicius Lima & Etienne Combrisson

The synchronization of the activity from distinct brain areas has been proposed to be one of the mechanisms by which them integrate while processing similar inputs in order to exchange information or encode the stimulus (Buzsaki, G., 2006; Fries, P., 2015). Based on this hypothesis, the time-course of the functional connectivity (dFC) can be measured from brain signals using metrics that capture their phase-relation such as the cross-spectra, the phase-locking value (PLV), and the coherence (Bastos A.M., Schoffelen J.M. 2006).
Additionally, apart from estimating those metrics in a time-resolved manner, in order to be able to relate the dynamics of the phase-coupling and task-related behavioral events, it is also relevant to assess the dFC at single-trial level, hence, avoiding averaging out non-phase-locked bursts of synchronization that are present in the dFC and may correspond to brain states relevant to determining, for instance, whether the information is being encoded during cognitive tasks by the coordinated activity of multiple cortical areas.
Currently, xfrites - the testing repository associated to Frites - has a function that estimates dFC in terms of the aforementioned metrics. For the present project, we aim to integrate it with Frites and, more specifically, we aim to improve the documentation of the function, refine the current implementation and include code for unit testing. Other goals are to implement notebooks with examples that allow the user to have a better understanding of how to first, set the parameters to estimate the spectral connectivity and seconds, to interpret the metric's outcome, what are its advantages and drawbacks.
Keywords: communication through coherence; spectral analysis; wavelet coherence; dynamic functional connectivity.

The following points are going to be addressed in this project:
  • Describe the methods to other participants
  • Refine the method implementation to estimate spectral connectivity present in xfrites
  • Create the documentation for the method
  • Write smoke and functional unit tests
  • Create examples illustrating the purpose of the single-trial coherence / PLV

Required skills

This is a python project, the following skills are highly recommended:
Python programming 80%
Spectral analysis 50%

BIDS-ephys: let's BIDSify your animal electrophysiology data

by Sylvain Takerkart & Julia Sprenger (coordinators of the BIDS-extension proposal for animal electrophysiology)
Other members of the INCF working group on Neuroscience Data Structure

We are working on an extension proposal so that BIDS (a standardized data structure described here) can support electrophysiological data recorded in animal models.
Our working document that describes the data organization for animal ephys data is available here. It is now ready to be used, and we propose here that electrophysiologists can attempt applying this organization to their data, with the help of the community.

The following points are going to be addressed in this project:
  • Have as many electrophysiologists as possible convert their data to the proposed BIDS-ephys organization
  • Identify potential problems in the specifications, solve them
  • Make progress on the tools required to facilitate this conversion (some of them being present here)

Required skills

This is a project where coding is not required. If you know or have electrophysiology data recorded in animal models, come over and tell us about it! Some other project participants might do some coding, in Python. Appreciated skills:
Expertise in electrophysiology 80%
BIDS enthusiasts / expert 70%
Python programming 60%

BrainSpread: Misfolded proteins spreading

by Alessandro Crimi, Luca Gherardini & Aleksandra Pestka

Most of neurodegenerative diseases are believed to be related to the spreading of misfolded proteins. Controversies still arise whether which path those proteins follow, and whether brain structures are damaged during this spreading or they were predisposing it.
In the project, DWI data collected from Alzheimer's patients is used. The data flow includes: preprocessing, tractography generation, connectivity matrix calculation and connectivity graph visualization. The next step is to simulate a spreading of misfolded proteins based on generated connectivity matrix and the location of seeds for Alzheimer's beta-amyloid in the brain.

The following points are going to be addressed in this project:
  • Discuss and code the simulation methods (heat kernel based diffusion, epidemic simulation model etc.).
  • Discuss the problem of a connectome structure change in time caused by neurodegeneration

Required skills

This project requires a mix of coding and non-coding skills:
Knowledge of neuroimaging, simulation modeling 90%
Python programming 70%

Going beyond pairwise interactions by digging into Higher-Order Interactions

by Etienne Combrisson, Andrea Brovelli & Daniele Marinazzo

Modern theories suggest that cognitive functions emerge from the dynamic coordination of neural activity over large-scale and hierarchical networks. Currently, the characterization of a network and therefore, the functional interactions between brain regions, is usually performed using metrics of Functional Connectivity (FC). FC analysis is mostly based on the quantification of statistical relations between pairs of brain regions. However, pairwise interactions are probably insufficient to explain the emergence of more complex brain network interactions, such as during goal-directed learning tasks. Here, we propose to move beyond pairwise interactions by studying at Higher Order Interactions (HOI) i.e. quantifying the information carried by groups of "over-two" brain regions (= multiplets). As a first step, a framework called O-information (= Information about Organizational structure) was recently proposed to characterize redundancy- and synergy-dominated systems. This framework has recently been extended with the dOinfo (= dynamic Information about Organizational structure) to quantify how multiplets of variables carry information about the future of the dynamical system they belong to. This dOinfo extension allows to separate multiplets of variables which influence synergistically the future of the system from redundant multiplets.
Since (d)OInfo frameworks are recent, the math underneath are quite new and we are not necessary familiar with it. The overall goal of this project is to understand the methods by looking at the reference papers and the Python / Matlab implementations of both (d)OInfo.

The following points are going to be addressed in this project:
  • Go through the reference papers (i.e. Rosas et al. 2019 and Stramaglia et al. 2021) to build an intuition of the math undergoing the HOI
  • Go through the Python toolbox HOI_toolbox to understand what are the input / output types, to identify the main accessible functions such as understanding the internals
  • Make the package easy to install, probably clean up so files
  • Identify whether there are coding bottlenecks that could be easily solved to speed up computations (soft Numba, multi-core, tensor-computations etc.)
If we still have time, here are some new features that could be added to the Python toolbox:
  • Implementation of false discovery rate for the significance of the multiplets
  • Speed up of the bootstrap
  • Include some plotting functions

Required skills

This project requires a mix of coding and non-coding skills:
Computational Neuroscience 70%
Python programming 70%
Information theory 60%
Mathematics 50%
Matlab 30%

Macapype: external pipelines

by David Meunier, Kepkee Loh, Julien Sein, Bastien Cagna & Olivier Coulon

Macapype is a python package based on nipype, wrapping the tools required for NHP MRI anatomical segmentation. Macapype provide several processing pipelines, allowing optimisation and adaptation, depending on the quality of the images (SNR), acquisition sequences and antenna types, as well as species.
Today, macapype is provided as a github repo, a pip install, and a docker/singularity image. The accepted inputs have to be in BIDS format, and macapype pipelines are callable thanks to a command line interface (CLI).
The following points are going to be addressed in this project:
Macapype is mostly oriented to PNH anatomical segmentation. However, many applications can be included starting from a good segmentation quality :
  • ACT (anatomically contrained tractography) from diffusion data
  • Mesh and surfaces generation that are not pure PNH but that are useful to be linked to macapype segementation
We would like to discuss the possibility of integration o this tools as external pipelines, i.e. not directly part of macapype, but that can be added in global workflows based on nipype.

Required skills

This is a Python project, coding skills and imaging knowledge are required:
Python programming 85%
PNH anatomical MRI processing 75%
Nipype 60%

Development of the SLAM -Surface anaLysis And Modeling- python package

by Alexandre Pron, Guillaume Auzias & the MeCA team

Tasks and objectives we would like to achieve can be found in the brain-slam github project.
Slam is an open source python package dedicated to the representation of neuroanatomical surfaces stemming from MRI data in the form of triangular meshes and to their processing and analysis.
Main features include read/write gifti (and nifti) file format, geodesic distance computation, several implementations of graph Laplacian and Gradient, mesh surgery (boundary identification, large hole closing), several types of mapping between the mesh and a sphere, a disc… Have a look at the examples on the documentation website.

The following points are going to be addressed in this project:
  • Help potential users to get familiar with this python package. The first BrainHack day could be considered as a training session for slam users
  • Further improve the documentation
  • Further improve the code quality of slam 's core functions (code linting, docstrings, refactoring, utils module)
  • Further improve code quality with new unitest and potential speed-up of specific pieces of code such as for instance the computation of the curvature
  • Brainstorm about slam perimeter
  • Design a logo
We are of course also open if new features are proposed from the contributors.

Required skills

This is a Python project, enthusiasm and coding skills are required:
Share ideas and good time 100%
Python programming 50%
Mesh processing 20%

Tools for MRI diffusion in human and NHP

by Julien Sein, David Meunier, Arnaud Le Troter & Hugo Dary

There are many pipelines for processing diffusion MRI data (diffuse, qsiprep, mrtrix pipelines, FSL pipelines, dipy, Designer, etc.). For newcomers in the field of Diffusion MRI, it may be overwhelming to pick the tools that best fits the needs, as well as to define the needs themselves. (NHP specific) : NHP image processing has specific challenges such as image reorientation, which can be more tricky when the Diffusion Vectors as to follow the transformation applied to the diffusion images.

The following points are going to be addressed in this project:
  • We aim at sharing expertise, identify specific problems encountered by users, and define quality metrics (quality of orientation vectors, quality check)
  • (NHP specific) : We also aim at integrating segmentations computed externally to mrtrix and FSL pipelines, that are non-specific pipelines (i.e. also available for human images)

Required skills

For this project general knowledges about MRI pipelines are recommended:
Building pipelines 70%
MRI diffusion 50%
MRI processing 50%


Sylvain Takerkart

Research Engineer

David Meunier

Research Engineer

Dipankar Bachar

Research Engineer

Julia Sprenger


Martin Szinte


Alexandre Pron

Research Engineer

Simon Moré


Ruggero Basanisi

PhD Student

Etienne Combrisson




Faculté de Médecine, 27 Boulevard Jean Moulin, 13005 Marseille