Project Leads:
• Amonida Zadissa, UK DRI, IO-FAST, (project lead, FAIR expert, sub-project lead P2)
• Samual Jackson, UK DRI, IO-FAST, (project lead, FAIR expert)
• Jan-Philipp Mallm, DKFZ, IO-FAST, (project lead, spatia-omics expert,sub-project lead P1)
• Teresa Zulueta Coarasa, EBI (BIA expert, sub-project lead P1)
• Aybuke Yoldas, EBI (BIA expert, sub-project lead P3)
• Naveed Ishaque, BIH (co-organiser, spatial-omics expert)
• Pavankumar Videm, Freiburg (Galaxy and spatial-omics expert, sub-project lead P4)
• Amirhossein Naghsh Nilchi, Freiburg (Galaxy and spatial-omics expert, sub-project lead P4)
• Luca Marconato (SpatialData expert, sub-project lead P2+P3)
For questions concerning the project please contact
Overview
This project unites the interoperability-focused IO-FAST consortium and the community-driven SpaceHack initiative to address challenges regarding spatial omics data. We propose four aligned subprojects to enhance metadata standards, data conversion, and visualization/upload tools for spatial omics. In addition, we will address sharing analytical pipelines, as well as the pre- and post-processing steps, to ensure reproducibility and thereby directly supporting open science and infrastructure goals of de.NBI/ELIXIR Germany.
In particular, this hackathon aims to interface “SpatialData” objects, the emerging community-accepted data format/container for spatial omics data, with the EBI BioImageArchive (BIA) and the Galaxy ecosystem to enhance FAIRness of data and data reusability from the outset of a project.
Objectives
- Advance Metadata Standards and Ontologies
- Streamline Data Conversion
- Enable Advanced Data Sharing and Visualization
- Implementation of Interoperable Datastructures into Galaxy Workflows
Subprojects
- Metadata validator and ontologies
Continuing from the previous SpaceHack hackathon, develop a metadata model and validation of correct use of ontologies. Feedback welcome from both experimentalists and bioinformaticians when it comes to the practical implementation of recording and using metadata. Based on the results, prepare training material to disseminate best practices in regard to metadata acquisition and BIA upload (aligned with subproject 3).
- Conversion of “historical” data from repository into SpatialData
Streamline the conversion of data from repositories into SpatialData objects. Standardizes handling, reading and analyzing data - while maintaining metadata information as stringently as possible. This of course also applies to data made available by collaboration partners or commercial providers.
- Pipeline to upload and share SpatialData via BIA
So far it is not possible to directly upload and share SpatialData objects via the BIA. While galleries exist that give a glimpse of the data sets, data visualization of e.g. molecule positions or cell types is not yet possible. The first step to provide this functionality is the digestion of SpatialData objects which will be addressed in this subproject. Visualization of highly complex datasets can be achieved via Vitessce and thus a prototype will be developed.
- Training materials/workflows hackathon as part of Spatial2Galaxy
Enhancing the integration of spatial transcriptomics within the Galaxy ecosystem, we will develop training materials and workflows for SpatialData analysis. It will focus on efficient retrieval of data from and push data to the BIA, while also ensuring that Galaxy is fully compatible with SpatialData objects by introducing a new datatype and sniff function. Additionally, establish best practices for spatial data handling and create workflows that guide users from raw data through to downstream analysis. Exporting via RO-Crates will be added as an additional asset for sharing data with a broader community. Similar to project three Vitessce will be used as a powerful tool for data visualization.