A training school for the new generation of bioimage analysts. Topics: workflow-based image analysis and new integrated methods for cloud computing applied to life sciences.
https://docs.google.com/forms/d/e/1FAIpQLSeyiFfdfRYAcQQB73dGy7WuclvkB_RiKxHncAd_HGwN_geFgA/viewform
| Date | Day | Topics | Sessions | Time |
|---|---|---|---|---|
| 23 September | Day 1 | Introduction to bioimage analysis, tools, and workflows | “BioImage: correlated multimodal imaging in life sciences and the problem of big data management for core facilities” “Introduction to what is bioimage analysis” “Image Data Services at Euro-BioImaging: Community efforts towards FAIR Image Data and Analysis Services” “Jupyter for interactive cloud computing” “Jupyter exercises” |
14:00 CEST 15:00 CEST 15:30 CEST 16:30 CEST |
| 30 September | Day 2 | FAIR principles Cloud hosted image data and cloud infrastructures Machine and Deep Learning on the cloud I Machine and Deep Learning on the cloud II |
“Introduction to FAIR principles for computational workflows” “Cloud hosted image data storage, visualisation and sharing”
“Machine and Deep Learning on the cloud: Segmentation” |
14:30 CEST 16:00 CEST 17:30 CEST |
| 7 October | Day 3 | Examples of bioimage analysis workflows |
“Zero code deep-learning tools for bioimage analysis” “CellProfiler for HCS data on the cloud” “Analysis of Microtubule Orientation” |
14:00 CEST 15:30 CEST |
| 14 October | Day 4 | Parallelization: from CPU to GPU - how to speed up workflows Work on your own data |
“Parallelization and heterogeneous computing: from pure CPU to GPU-accelerated image processing” “Introduction to Galaxy workflow environment” Trainees will be split into groups and they will start working on the selected workflows to convert them into a cloud solution |
14:00 CEST 15:30 CEST |
| 21 October | Day 5 | Benchmarking theory Benchmarking tools Work on your own data |
“Metrics and Benchmarking” “BIAFLOWS: A BioImage Analysis workflows benchmarking platform” Trainees will present their cloud workflows |
14:00 CEST 15:45 CEST |
Scientific Organizers: Rocco D'Antuono, Paula Sampaio, Mafalda Sousa, Clara Prats, Marion Loveaux.
Scientific Advisory Board: Julien Colombelli, Paula Sampaio, Julia Fernandez-Rodriguez, Gaby Martins.
Event Timing: Training school distributed over 5 CEST Friday afternoon classes on September 23rd and 30th - October 7th, 14th, and 21st.
Event type: Online.
Capacity: max 25 participants.
Participation fee: Euro 100 (to be paid after admission).
Registration form: https://docs.google.com/forms/d/e/1FAIpQLSeyiFfdfRYAcQQB73dGy7WuclvkB_RiKxHncAd_HGwN_geFgA/viewform
Registration deadline: 14 August 2022
Contacts: [email protected]
Please use the following subject: Defragmentation TS
Day 1:
a) before the TS starts the trainees are pleased asked to register an account for https://usegalaxy.eu/login and optionally a GitHub account: https://github.com/
Day 2:
a) before Day 2 the trainees need to have a working Google account (not a linked account that you manage with Google) to run Colab notebooks: https://colab.research.google.com/notebook . The Google account should have enough free space (~ several GB) on it.
b) follow instructions from Dr. Christian Tischer to access BAND and run the script that installs the needed software on the remote desktop.
c) read the message from Dr. David Rousseau about slides and introduction to Google Colab for those who are not familiar yet with it.
Day 3:
a) it is required to have a Google Drive account and an instance of Fiji with the plugin deepImageJ.
b) download CellProfiler here: https://cellprofiler.org/releases . Good if trainees get familiar with CP beforehand by e.g. watching this tutorial video: https://www.youtube.com/watch?v=QrzHQIiIDKM .
Day 4:
a) trainees can try the devbio-napari plugin collection before the lesson: https://github.com/haesleinhuepf/devbio-napari#installation.
Day 5:
a) Trainees are kindly asked to complete the presentation for the session "Work on your own data".
b) trainees have to register a GitHub account before the lesson: https://github.com/.
c) during the lesson trainees will be asked to register a Docker Hub account with name corresponding to the GitHub one: https://hub.docker.com/.
- “BioImage: correlated multimodal imaging in life sciences and the problem of big data management for core facilities”
Julia Fernandez-Rodriguez - “Introduction to what is bioimage analysis”
Kota Miura - “Image Data Services at Euro-BioImaging: Community efforts towards FAIR Image Data and Analysis Services”
Aastha Mathur - “Jupyter for interactive cloud computing”
Guillaume Witz
- “Introduction to FAIR principles for computational workflows”
Paper: Software Citation Principles
YouTube: REMBI
UK Conference of Bioinformatics and Computational Biology
Carole Goble - “Cloud hosted image data storage, visualisation and sharing”
Installation
Christian Tischer - “Machine and Deep Learning on the cloud: Classification”
Get familiar with Google Colab
David Rousseau - “Machine and Deep Learning on the cloud: Segmentation”
Ignacio Arganda-Carreras
- “Zero code deep-learning tools for bioimage analysis”
Daniel Sage - “CellProfiler for HCS data on the cloud”
Input Modules Tutorial
Preprint: Distributed-Something: scripts to leverage AWS storage and computing for distributed workflows at scale
Beth Cimini and Anna Klemm - “Analysis of Microtubule Orientation”
Thomas Pengo
- "Parallelization and heterogeneous computing: from pure CPU to GPU-accelerated image processing”
Robert Haase - “Introduction to Galaxy workflow environment”
Beatriz Serrano-Solano and Björn Grüning
- “Metrics and Benchmarking”
Michal Kozubek - “BIAFLOWS: A BioImage Analysis workflows benchmarking platform”
"Adding a workflow to BIAFLOWS"
BIAflows
Sébastien Tosi, Volker Baecker, and Benjamin Pavie