Data management, organization and analysis training

Bringing in silico and in vitro protocols together with best cell culture practices and harmonised data management provides a sound foundation for a reproducible evidence-based science for animal-free testing of liver toxicity. The proposed training will offer solutions to:

  • uploading, harmonisation, annotation and storage of data and metadata,
  • storage aligned to FAIR data principles,
  • user-friendly, non-expert level searching and browsing, as well as analysis and visualisation, both interactively and in scripting mode,
  • options for customisation to address specific project requirements including optimal interplay with other existing and newly developed infrastructure.

Furthermore, the solutions will be illustrated by dedicated sessions on:

  • Managing Project Data
  • Organising Laboratory Data Workflows
  • Predictive (toxicology) modelling workflows

As such, knowledge on how to integrate and harmonise project-driven data involving multiple users ( including cases when data is available in different file formats) will be gained. Also, the participants will learn how to store raw or processed laboratory data, avoid errors due to manual data transfer and last but not least how to select and combine data to make predictions by using advanced approaches available this area.

Target audience: people responsible for data collection, storage, analysis and interpretation.

Provisional program

Day 1

8.30h - 9.00h The FAIR data principles: introduction
9.00h - 9.45h The art of managing project data: comprehensive, flexible data management environment to collect, analyse, visualise, share and store this data and metadata.
9.45h - 10.00h Break
10.00h - 12.00h Hands-on session:
- The example from EU-ToxRisk
- Customized data
12.00h - 13.00h Lunch break
13.00h - 14.00h Introduction to (semi-automated) laboratory data workflows
- Examples from the bench
14.00h - 17.00h Hands-on session:
- Analysis of toxicogenomic data (calculation of fold changes, adding gene annotations)
- Linking of toxicogenomic and protein binding data - example of AOPLink case study from OpenRiskNet
- Building a predictive model for a particular endpoint
- Integrating outcomes from multiple predictive models into a consensus prediction - example of ModelRX case study from OpenRiskNet.