NEWS

---

omiprep 0.9.0

First public, peer-review release. omiprep is the successor to metaboprep, generalised from Nightingale/Metabolon metabolomics to mixed metabolomics and proteomics platforms.

Release

• 2026-06-26: Version 0.9.0 formally released as a pre-release on GitHub.
• Archived on Zenodo with a package DOI: 10.5281/zenodo.20941283 (concept DOI, resolves to the latest version).

Data import

• Unified readers into a single Omiprep object: read_metabolon() (v1.1, v1.2, and v2 multi-tab formats), read_nightingale() (single- and multi-sheet formats), read_olink(), and read_somalogic().
• Synthetic example datasets bundled for the Metabolon and Nightingale formats.

Quality control

• quality_control() orchestrates sample/feature missingness filtering, total-sum-abundance outliers, PC-based sample outliers, feature-tree reduction, and winsorisation, recording all exclusions on the object.
• Building blocks exported individually: missingness(), outlier_detection(), pc_and_outliers(), total_sum_abundance(), tree_and_independent_features(), feature_summary(), sample_summary(), summarise().

Normalisation & layers

• batch_normalise() for platform / run-mode batch normalisation.
• Multi-layer Omiprep@data array with add_layer().

Reporting & export

• generate_report() renders HTML/PDF QC reports (available_report_templates()).
• export_omiprep(), export_metaboanalyst(), and export_comets() for downstream tools.
• shiny_app() interactive explorer.

Compatibility

• run_metaboprep1() preserves the original metaboprep workflow.