EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

UniCoracle: automated hierarchical feature selection via bottom-up propagation and top-down skimming using the UniCorP algorithm and the Coracle machine-learning framework

Sebastian Staab iD, Anny Cardénas iD, Raquel S Peixoto iD, Falk Schreiber iD, Christian R Voolstra iD

Bioinformatics · 2026

DOI: 10.1093/bioinformatics/btag507 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Sebastian Staab und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biology, University of Konstanz, Konstanz 78457, Germany. Hauptthemen laut MEDLINE: Machine Learning, Algorithms, Microbiota, Computational Biology. Die Arbeit nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: King Abdullah University of Science and Technology (KAUST), the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Summary Identifying meaningful associations between microbial communities and measured physiological or environmental variables becomes increasingly complex and computationally demanding given the continuous growth of microbiome datasets. The Coracle machine learning (ML) framework was recently developed to address this issue by integrating multiple data transformations, feature selection techniques, and ML models to yield condensed lists of features that align to target variables of interest. Further, we recently developed the UniCorP feature aggregation algorithm to identify uniquely correlated features (UNICORNs) based on the UniCor metric that iteratively enrich each taxonomic level in an automated bottom-up approach. Here we present UniCoracle, a fully automated analytical framework that integrates UniCorP’s bottom-up propagation approach with a subsequent and newly developed top-down skimming (TDS) strategy, implemented with the Coracle ML framework. This combined approach leverages the inherent taxonomic structure of microbiome community data (e.g., ASVs derived from 16S rRNA gene amplicon sequencing data) to maintain predictive stability, reduce computational runtime, and identify biologically meaningful taxonomic associations. We compare the original, non-hierarchical Coracle with the TDS Coracle method and the UniCoracle approach. Evaluations across the tested datasets show that UniCoracle achieves competitive or improved predictive performance relative to both Coracle’s multi-step and the TDS-based Coracle implementations and demonstrate UniCoracle’s improvements in predictive accuracy over both methods. UniCoracle provides full control over feature set size and runtime, offering a streamlined and user-friendly framework for biological hypothesis generation. It identifies features (e.g., bacterial taxa) at the lowest (most specific) hierarchical level (e.g., ASV or species within a taxonomic hierarchy) that are associated with continuous target variables. Availability UniCoracle is freely accessible via a dedicated web server at micportal.org. The source code is open source and available on GitHub at github.com/SebastianStaab/UniCoracle.git and Zenodo at https://doi.org/10.5281/zenodo.19050205.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningAlgorithmsMicrobiotaComputational BiologySoftwareRNA, Ribosomal, 16S

Substanzen

RNA, Ribosomal, 16S

Autor:innen und Institutionen

  • Department of Biology, University of Konstanz, Konstanz 78457, Germany.
  • Department of Biology, American University, Washington, DC, United States.
  • Red Sea Research Center (RSRC), Biological and Environmental Sciences and Engineering Division (BESE), King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia.
  • Department of Computer and Information Science, University of Konstanz, Konstanz 78457, Germany.
  • Faculty of Information Technology, Monash University, Clayton, Victoria 3800, Australia.

ORCID-Profile: Sebastian Staab, Anny Cardénas, Raquel S Peixoto, Falk Schreiber, Christian R Voolstra

Förderung

  • King Abdullah University of Science and Technology (KAUST)
  • the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)

Zitieren

APA

Staab, S., Cardénas, A., Peixoto, R. S., Schreiber, F., & Voolstra, C. R. (2026). UniCoracle: automated hierarchical feature selection via bottom-up propagation and top-down skimming using the UniCorP algorithm and the Coracle machine-learning framework. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag507

Vancouver

Staab S, Cardénas A, Peixoto RS, Schreiber F, Voolstra CR. UniCoracle: automated hierarchical feature selection via bottom-up propagation and top-down skimming using the UniCorP algorithm and the Coracle machine-learning framework. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag507

BibTeX
@article{Staab2026,
  author = {Staab, Sebastian and Cardénas, Anny and Peixoto, Raquel S and Schreiber, Falk and Voolstra, Christian R},
  title = {UniCoracle: automated hierarchical feature selection via bottom-up propagation and top-down skimming using the UniCorP algorithm and the Coracle machine-learning framework},
  journal = {Bioinformatics},
  year = {2026},
  doi = {10.1093/bioinformatics/btag507}
}
Kostenlos starten →