Skip to end of metadata
Go to start of metadata

Overview

On request from the institute of biochemistry and the institute of microbiology of D-BIOL, SyBIT will offer a half day workshop about openBIS in the morning of Thursday, April 15th on Hönggerberg. The workshop will have two parts:

  • Part 1 is for interested parties who would like to get an overview about what the system can do for them. There will be three short presentations & demos given by expert users of labs in IMSB and LMC who use the system.
  • Part 2 will be an opportunity for expert users to talk about where they would like the system to go, what they would like to see improved and new use cases that they would like to see covered.

Location and time

Time: Thursday, April 15th, 9:00 - 12:30

Location: Hönggerberg, HPM C51

Agenda

Time

Topic

Speaker

9:00

Welcome

Peter Kunszt (SyBIT, SystemsX.ch)

9:05

Short introduction to openBIS

Bernd Rinn (C-ISD, D-BSSE)

 

Part 1: current use: presentations, demos, questions

 

9:25

openBIS Proteomics & p-grade: an integrated system for proteomics data analysis and management

Lars Malmstroem (Aebersold Lab, IMSB, D-BIOL)

9:45

openBIS Metabolomics: supporting the Sauer lab's Metabolomics massive injection and quantification workflows

Jörg Büscher (Sauer Lab, IMSB, D-BIOL)

10:05

openBIS Screening & HCDC: towards an integrated pipeline for High Content Screening analysis

Karol Kozak (LMC, D-BIOL) / Tomasz Pylak (C-ISD, D-BSSE)

10:30

Coffee break

 

 

Part 2: Ideas for improvements, issues to be fixed, new use cases

 

11:00

Moderated open discussion

all

Minutes

Peter Kunszt (SyBIT) opened the meeting with giving an overview about SyBIT's vision and mission and how openBIS fits into it. There were questions about whether SyBIT will be able to reach its goal to provide the researchers with long-term sustainable solutions. Peter explained that SyBIT is now well prepared to take on the challenge, it has achievements to show and, given there is further support by SystemsX.ch for the project, will build on them.

Bernd Rinn (CISD) gave an overview about what openBIS is and what its main use cases are. He stressed the point that openBIS has a generic part and can be used out of the box, but that it is meant to be customized and extended to understand specific types of data and integrate into the workflow of a specific lab or group of labs and that it has been built with extensibility in mind. Bernd explains about the core structures of openBIS, the extension points and the different levels of integration into measurement and analysis workflows. Bernd explained also how to get data and meta data into and out of the system and what are the options to integrate it with existing data analysis pipelines. During the talk, discussions spun off on what is the best way to ensure that meta data actually end up in the system and how the system can be made to play nicely with hierarchical storage systems and staging approaches.

Lars Malmstroem (IMSB) showed how openBIS and p-grade (a middleware system) is used in the new (still alpha-stage) PhosphoNetX / Aebersold integrated Proteomics workflow. He emphasized the need for a tight integration between openBIS, the p-grade middleware server, the conversion nodes and the (multi-tier) storage infrastructure to cover the Aebersold proteomics pipeline starting from the mass spec device to the proteomics search results (using a set of proteomics engines). Lars gave a live demo of the system, showing how biologists can query the system for the results of protein search experiments.

Joerg Buescher (IMSB) showed how openBIS is used to manage metabolomics data for the YeastX and LiverX projects. Raw data and result data for the Sauer lab's high throughput and quantification workflows are managed in openBIS. A custom database schema is automatically populated by openBIS which is then queried in the matlab based analysis pipeline which finally feeds its results back into openBIS. Joerg gives a quick live demo and shows the special reporting operations available for the quantification data sets. openBIS is used productively in the Sauer lab since 9 months and contains approximately 44000 datasets with about 140 Gigabyte of data.

Tomasz Pylak (CISD) showed the work being done in openBIS to support High Content Screening data management. He gave a live demo of openBIS for Screening on demo data from LMC, showing many of the advanced features like merging channels on an image, linking images to feature vectors and showing all images where a certain gene is expressed in an siRNAi screen. He explained the concept of openBIS' remote APIs and how they can be used to connect openBIS with data analysis applications. Karol Kozcak (LMC) showed how this API has been used to write a KNIME reader node for openBIS data and then process the images read from there with the HCDC nodes.

The workshop was well received and discussions started on how the system can be used for other data like FACS data that groups in the institute of microbiology have to find a good data management solution for. These discussions will have to be followed up on in a separate meeting.

  • No labels