EOSC-Life Common provenance model for processing biological material, data generation and computational workflows

The exchange of research data and physical specimens has become an issue of major importance for modern research. Many reports indicate problems with quality, trustworthiness, and reproducibility of research results, mainly due to poor documentation of the data generation or the collection of specimens. The significant impact of flawed research results on health, economics, and political decisions has frequently been stated. Consequently, professional societies and research initiatives call for improved and standardised documentation of the data and specimens used in research studies. Provenance information is information about the history of an object, and depending on its content, it can be used to verify the quality and reliability of research results. In particular, provenance has been adopted in scientific domains to support a traceable lineage of research objects, such as biological material, data, or workflows. The purpose of provenance information is not to replace existing logging infrastructures or information systems that currently handle documentation of research objects and related processes or their metadata. Provenance information serves as long-term available information that can be typically used to assess the quality of the documented object (fitness for purpose) or to reproduce documented processes. As the research objects are typically exchanged between organisations, each organisation can provide provenance information only about a portion of the documented research object’s life cycle. WP6 in the EOSC-Life Project aimed to design a paradigm to capture and exchange interoperable provenance information, creating a common ground across different domains in the Life Sciences and building a harmonised understanding of provenance information. The work led to the definition of the Common Provenance Model (CPM), described in Deliverable 6.2, starting with the design of components of distributed provenance information to enable interlinking of provenance information generated in different organisations involved in the research process, such as biobanks, research centres, universities, or analytical laboratories. As detailed in Deliverable 6.2, this distributed provenance information model builds on an existing provenance information standard, W3C PROV, and follows a general provenance composition pattern. This document summarises the evolution of the work previously described in D6.2 and presents the CPM for processing biological material, data generation, and computational workflows. The CPM is designed to address the reproducibility issues in sciences by supporting the traceability of documented research objects in multi-organisational environments. This document is focused on, but not limited to, provenance of research objects, such as datasets, biological or environmental specimens, computational models, software, tools, or experimental results. The CPM supports reproducibility by providing a conceptual foundation for trustworthy provenance information, where each of its parts is generated, stored, and managed by a different organisation. In particular, the CPM defines a common horizontal framework to represent links between the provenance parts, and to represent common semantics to bind domain-specific provenance information. The CPM serves as an open conceptual foundation for the development of the ISO 23494 – Provenance information model for biological material and data standard series