Orchestrating and sharing large multimodal data for transparent and reproducible research

Area of Science:

• Biomedical Data Science
• Computational Biology
• Open Science

Background:

• Reproducibility is critical for scientific validity and open science.
• The increasing complexity and volume of biomedical data hinder its processing, analysis, and sharing.
• Current data management practices struggle to keep pace with FAIR (findable, accessible, interoperable, and reusable) principles.

Purpose of the Study:

• To develop a flexible, cloud-based framework for reproducible processing of multimodal biomedical data.
• To address the challenges of handling complex and growing datasets in cancer research.
• To enhance the findability, accessibility, interoperability, and reusability of biomedical data.

Main Methods:

• Development of ORCESTRA, a cloud-based platform for automated, user-customizable processing pipelines.
• Integration of clinical, genomic, and perturbation data from cancer samples.
• Implementation of features for creating documented data objects with persistent identifiers (DOI) and version management.

Main Results:

• ORCESTRA provides a flexible framework for reproducible processing of multimodal biomedical data.
• The platform enables automated and customizable analysis of diverse cancer-related datasets.
• Integrated data objects with DOIs and version control facilitate data sharing and future studies.

Conclusions:

• ORCESTRA offers a robust solution for reproducible biomedical data processing and management.
• The platform supports open science principles by enhancing data sharing and reusability.
• ORCESTRA is poised to advance cancer research through improved data accessibility and integrity.