% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@INPROCEEDINGS{Khler:1046208,
      author       = {Köhler, Cristiano and Grün, Sonja and Denker, Michael},
      title        = {{S}upporting {FAIR} {P}rinciples in {D}ata {A}nalysis
                      {T}hrough {S}emantically-{E}nriched {P}rovenance},
      reportid     = {FZJ-2025-03744},
      year         = {2025},
      abstract     = {Scripts that read input datasets and generate result files
                      are frequently used to construct computational workflows for
                      the analysis of neural activity data obtained by
                      electrophysiology recordings [1]. The increased complexity
                      of datasets due to recent advances in recording techniques
                      are also associated with increased computational costs for
                      executing those workflows and generating analysis results.
                      Increasing the FAIR-ness [2] of electrophysiology data
                      analysis results will promote the efficient sharing of
                      results among collaborators or the research community. With
                      increased findability, a collaborator can easily access
                      specific results produced by complex analysis without
                      rerunning costly computations. If results are more
                      accessible, they are transparent and can be reused across
                      platforms and organizations. The increased interoperability
                      facilitates understanding specific analysis results despite
                      their generation by heterogeneous workflows, improving
                      collaboration and allowing the comparison of different
                      analysis results. Finally, reusable results allow
                      researchers to build on previous analyses, conserving
                      resources and speeding scientific discovery without
                      repeating complex computations. In this work, we investigate
                      an approach for describing the results generated by
                      electrophysiology data analysis workflows in order to
                      increase the FAIR-ness of results. We aim to go beyond
                      providing source codes and free-text descriptions to
                      facilitate querying, introspection and reuse of the results
                      by capturing and evaluating run-time provenance information.
                      We highlight several challenges within the workflows that
                      hinder the creation of such descriptions, including the
                      iterative characteristics of conventional analysis
                      scenarios, the absence of technical and semantic
                      standardization, and the presence of distinct software
                      implementations for existing analysis methods [3]. To
                      address those challenges, we first implemented Alpaca
                      (Automatic Lightweight Provenance Capture) as a framework to
                      generate machine-readable descriptions of the workflow
                      execution with minimal user intervention [4]. Alpaca
                      produces a detailed provenance record of the atomic analysis
                      steps represented by Python functions within workflow
                      scripts, that are serialized together with analysis results
                      using the W3C PROV standard [5]. Complementing the approach,
                      the provenance information can be enriched with semantic
                      information provided by ontologies. For workflows analyzing
                      electrophysiolgy datasets with recorded neural activity, we
                      implemented the Neuroelectrophysiology Analysis Ontology
                      (NEAO) to provide a unified vocabulary to standardize the
                      descriptions of the methods involved in the analysis of
                      extracellular electrophysiology data [6]. We demonstrate how
                      using NEAO to enrich the provenance captured by Alpaca helps
                      in describing analysis results produced by complex
                      real-world workflows for analyzing and comparing
                      heterogeneous data based on Elephant [7] and Cobrawap [8].
                      We highlight how the approach facilitates obtaining insights
                      on the results (e.g., using knowledge graphs), thereby
                      promoting the FAIR principles and facilitating sharing. We
                      also discuss extensions to other computational workflows
                      (e.g., neural simulation) and how the proposed approach may
                      help to also improve representing their results according to
                      the FAIR principles. REFERENCES [1] M. Denker et al.,
                      “Reproducibility and efficiency in handling complex
                      neurophysiological data,” Neuroforum, vol. 27, no. 1, pp.
                      27–34, Feb, 2021, doi:
                      https://doi.org/10.1515/nf-2020-0041 [2] M. D. Wilkinson et
                      al., “The FAIR Guiding Principles for scientific data
                      management and stewardship,” Sci Data, vol. 3, p. 160018,
                      Mar, 2016, doi: https://doi.org/10.1038/sdata.2016.18 [3] V.
                      A. Unakafova and A. Gail, “Comparing Open-Source Toolboxes
                      for Processing and Analysis of Spike and Local Field
                      Potentials Data,” Front Neuroinform, vol. 13, p. 57, Jul,
                      2019, doi: https://doi.org/10.3389/fninf.2019.00057 [4] C.
                      A. Köhler et al., “Facilitating the Sharing of
                      Electrophysiology Data Analysis Results Through In-Depth
                      Provenance Capture,” eNeuro, vol. 11, no. 6, p.
                      ENEURO.0476-23.2024, May, 2024, doi:
                      https://doi.org/10.1523/ENEURO.0476-23.2024 [5] P. Groth and
                      L. Moreau. “An Overview of the PROV Family of
                      Documents.” PROV-Overview.
                      https://www.w3.org/TR/prov-overview (accessed on 28 April
                      2025) [6] C. A. Köhler, S. Grün, and M. Denker.
                      “Improving data sharing and knowledge transfer via the
                      Neuroelectrophysiology Analysis Ontology (NEAO),”
                      arXiv:2412.05021, Dec, 2024, doi:
                      https://doi.org/10.48550/arXiv.2412.05021 [7] R. Gutzen et
                      al., “A modular and adaptable analysis pipeline to compare
                      slow cerebral rhythms across heterogeneous datasets,” Cell
                      Rep Methods, vol. 4, no. 1, p. 100681, Jan, 2024, doi:
                      https://doi.org/10.1016/j.crmeth.2023.100681},
      month         = {Aug},
      date          = {2025-08-26},
      organization  = {2nd Conference on Research Data
                       Infrastructure (CoRDI), Aachen
                       (Germany), 26 Aug 2025 - 28 Aug 2025},
      subtyp        = {After Call},
      keywords     = {FAIR (Other) / electrophysiology (Other) / data analysis
                      (Other) / computational workflow (Other) / Python (Other) /
                      provenance (Other) / ontology (Other)},
      cin          = {IAS-6 / INM-10},
      cid          = {I:(DE-Juel1)IAS-6-20130828 / I:(DE-Juel1)INM-10-20170113},
      pnm          = {5235 - Digitization of Neuroscience and User-Community
                      Building (POF4-523) / 5231 - Neuroscientific Foundations
                      (POF4-523) / HDS LEE - Helmholtz School for Data Science in
                      Life, Earth and Energy (HDS LEE) (HDS-LEE-20190612) / HBP
                      SGA3 - Human Brain Project Specific Grant Agreement 3
                      (945539) / EBRAINS 2.0 - EBRAINS 2.0: A Research
                      Infrastructure to Advance Neuroscience and Brain Health
                      (101147319) / JL SMHB - Joint Lab Supercomputing and
                      Modeling for the Human Brain (JL SMHB-2021-2027) /
                      Algorithms of Adaptive Behavior and their Neuronal
                      Implementation in Health and Disease (iBehave-20220812)},
      pid          = {G:(DE-HGF)POF4-5235 / G:(DE-HGF)POF4-5231 /
                      G:(DE-Juel1)HDS-LEE-20190612 / G:(EU-Grant)945539 /
                      G:(EU-Grant)101147319 / G:(DE-Juel1)JL SMHB-2021-2027 /
                      G:(DE-Juel-1)iBehave-20220812},
      typ          = {PUB:(DE-HGF)24},
      doi          = {10.5281/ZENODO.16736244},
      url          = {https://juser.fz-juelich.de/record/1046208},
}