Introduction to the iPOP-UP HPC cluster

Summary

More info can be found here.

Organizers

• Olivier Kirsh
• Julien Rey
• Magali Hennion

Introduction to the iPOP-UP HPC cluster

Summary

More info can be found here.

Organizers

• Alix Silvert
• Julien Rey
• Magali Hennion

Seamless interactive workflow hackathon

Summary

Seamless is a programming framework developed at the RPBS platform for creating interactive, interoperable workflows and web services. A hackathon will take place on December 15th as an opportunity to discover how interactive web services can be created. An example of a web server already using this technology can be seen here.

The subject of the hackathon will be an interactive Seamless workflow developed by the Guerois-Andreani team, aimed towards building multiple sequence co-alignments as inputs for AlphaFold2 for the structural modeling of protein complexes. The idea is that manual fine-tuning is essential for producing accurate structural models that can be used in e.g. drug design.

The first part of the hackathon is didactic. The basic building blocks of Seamless will be explained. Interactivity and interoperability (including federated deployment) will be discussed. We will start from an empty workflow and modify it interactively while it is running. Step-by-step, the existing co-alignment scripts will be added, generating a default interactive web page.

In the second part of the hackathon, we will improve upon the existing Seamless workflow. Areas of improvement are: making the interactive web page more intuitive, visualisation of the co-alignments in the browser, validation schemas for the input parameters, better use of taxon IDs in building the co-alignments, and/or more fluid integration with AlphaFold 3D modeling.

The languages of the hackathon will be: Python, bash and Javascript for programming; and English and French for human communication. Skill in at least one programming language and one human language above are required. Expertise with protein sequences and/or structures is desirable, but not essential.

Links

Github project of the hackathon, code, data, workflows and notebooks can be found here.

Organizers

• Sjoerd de Vries
• Pierre Tufféry
• Raphaël Guerois
• Jessica Andreani

Jupyter

Programme

Jour 1

• 9:30-12:00 : Introduction à Jupyter
  • Pourquoi Jupyter ? Les avantages et les inconvénients de l'informatique interactive.
  • Comment Jupyter fonctionne avec n'importe quel langage de programmation, pas seulement Python.
  • Explication du fonctionnement de l'interface du navigateur et de l'exécution des cahiers Jupyter.
  • Exemples de cahiers Jupyter (Python, R, C ++, Fortran) et intégration à la ligne de commande.
  • Expérience pratique de l'exécution de commandes et de cellules de cahiers simples.

• 12:00-13:00 : Déjeuner

• 13:00-17:30 : Intégration de Jupyter dans vos recherches quotidiennes (session pratique)
  • Intégration pratique de Jupyter aux outils de calcul existants. Les participants sont invités à apporter leurs propres scripts, programmes et données.
  • Génération rapide de graphiques simples.
  • Intégration de plusieurs langages de programmation et démarquage dans le même cahier.

Jour 2

• 9:00-11:30 : Jupyter et analyse de données interactive (session pratique)
  • Analyse interactive de données à l'aide de bibliothèques scientifiques Python (Numpy, Scipy).
  • Bibliothèques pour visualiser et tracer (matplotlib, vega-lite, nglview, ggplot).
  • Des widgets interactifs pour contrôler l'analyse et la visualisation en temps réel.
• 11:30-12:30 : Jupyter pour la publication et l'enseignement
  • Explication des principes FAIR.
  • Utiliser des cahiers comme publications scientifiques.
  • Lancer Jupyter de n’importe où, en utilisant simplement un navigateur.
  • Exemples de cahiers Jupyter dans des publications de recherche existantes.
  • Démonstration de l'utilisation de Jupyter dans l'enseignement.

• 12:30-13:30 : Déjeuner

• 13:30-17:00 : Science reproductible et collaborative (session pratique)
  • Utilisation de Jupyter avec des référentiels de données.
  • Utiliser Jupyter avec Github.
  • Utiliser Jupyter avec Docker (repo2docker, binder).
  • Introduction à Jupyterlab.

Protein-peptide interactions: peptide identification, binding prediction and design

Location

CECAM-FR-MOSER University Paris-Diderot Amphi Turing Bâtiment Sophie Germain 8 place Aurélie Nemours - 75013 Paris

Organizers

• Pierre Tuffery, UMRS 973, Univ. Paris Diderot, France
• Martin Zacharias, Technical University of Munich, Germany
• Angelita Rebollo, CIMI - Univ. Pierre et Marie Curie, France

Advanced statistical approaches in Structural Bioinformatics

PIA "bioinformatique", ANR "Bayesian Inference Paradigm - Biology in Processors".

Conference schedule

Tuesday, October 24th

• 13:30-13:45 Welcome address (M. Nilges, P. Tufféry)
• 13:45-14:15 Modeling Enzyme-Substrate complexes for Glycosidases (P. Coutinho)
• 14:15-14:45 Tuning the interval Branch-and-Prune (iBP) algorithm for practical protein structure determination (B. Worley)
• 14:45-15:15 Potentials of mean force: from hack to math (T. Hamelryck)

• 15:15-15:30 Coffee break

• 15:30-16:00 Principle of Proteins Ensemble determination (C. Camilloni)
• 16:00-16:30 Modeling of structure and design of binding of peptide-protein interactions using Rosetta FlexPepDock (O. Furman)
• 16:30-17:00 Peptide modeling in isolation and in interaction : a probabilistic framework (P. Tufféry)

Wednesday, October 25th

• 09:30-10:00 Progress in Normal Modes refinement and generation of transition paths trajectories (M. Delarue)
• 10:00-10:30 From a conformer to another: the case of large amplitude conformational changes (YH. Sanejouand)

• 10:30-11:00 Coffee break

• 11:00-11:30 Evolutionary guided structural modelling of protein complexes (R. Guerois)
• 11:30-12:00 Bayesian modeling of biomolecular assemblies (M. Habeck)
• 12:00-12:30 Exploring biomolecular association using advanced sampling methods (M . Zacharias)

• 12:30-14:00 Lunch

• 14:00-14:30 Very large virtual library screening for ligand discovery (J. Irwin)
• 14:30-15:00 Mining ligand structural data from the PDB (H. Xhaard)
• 15:00-15:30 Mining target structural data and their druggability (AC. Camproux)

• 15:30-16:00 Coffee break

• 16:00-16:30 Sequence and 3D structure mining in rare genetics diseases (O. Poch)
• 16:30-17:00 Integrative modelling of transcriptional regulators (A. Dejaegere) (cancelled)

Location

The meeting will take place Amphitheater Turing, in the Sophie Germain building of University Paris Diderot. More information on the location and venue below.

IFB Bioinformatique Structurale: Workshop Web Viewers

Programme

• 10:30 A Multiple Sequence Alignement experience with a BioJS JavaScript viewer (A. Kress, Univ. Strasbourg)
• 11:00 Web, Interactive, 3D molecular visualization using JavaScript viewers (PV, NGL, JSmol, 3Dmol): an update (J. Rey, Univ. Paris Diderot)
• 11:30 An experience in interfacing PV and Galaxy (D. Ritchie/A. Chemardin, LORIA)

• 12:30-13:30 Déjeuner

• 14:00 Plotly: interactive plots using JavaSCript (S. de Vries, Univ. Paris Diderot)
• 14:30 Visualisation et édition de polymères ayant des structures non linéairesi (Maude Pupin, Univ. Lille)
• 15:15 UnityMol and WebGL (Xavier Martinez, Univ. Paris Sud)

• 16:00 End of the workshop