Skip to contents

Changelog

Source: NEWS.md

mrmhub 0.9.2

New features

mzTab-M import & export

  • New save_dataset_mztab() writes a processed MRMhubExperiment to the HUPO-PSI mzTab-M 2.0.0-M standard format (the format expected by repositories such as MetaboLights). Each analysis becomes an assay, each feature a Small Molecule Feature (SMF) row grouped into Small Molecule Summary (SML) rows by analyte. The abundance variable is selectable via variable (default "conc", with fall-back to raw intensity for unquantified data), and the metadata header can be enriched with instrument, contact and publication. Output is validator-passing against the official LIFS mzTab-M validator.

  • New import_data_mztab() imports mzTab-M files produced by other tools (e.g. Lipid Data Analyzer, MS-DIAL, MZmine). Each SMF becomes a feature and each assay an analysis; per-assay abundances are imported as feature_intensity, with feature identities and study_variable grouping (as batch_id) carried over where available.

  • Both directions are pure R and add no runtime dependency. See the Import & Export mzTab-M recipe article.

Handling of analytical sequence gaps in plot_runscatter() and plot_rla_boxplot()

These functions gain the following new arguments to handle gaps in the analytical sequence:

  • remove_gaps — if TRUE, gaps in the data (mainly caused by unannotated analysis results) are removed and indicated by vertical lines and index labels. The gap lines can be customised via gap_line_color, gap_line_width, gap_label_size, and gap_scale.

  • collapse_excluded — if TRUE, gaps in the x-axis caused by QC types that were not selected are collapsed. This is useful when plotting only samples or a subset of QC types.

Support for long feature names in plot_runscatter() and plot_responsecurves()

  • label_wrap and label_wrap_width allow long feature_id strip labels to be wrapped across multiple lines. Set label_wrap = TRUE to enable wrapping; use label_wrap_width to control the maximum line width in characters (default: 25).

Plotting multiple response curves per feature in plot_responsecurves()

plot_responsecurves() gains the following new arguments:

  • curve_layout — one of "overlay", "rows", or "cols". "overlay" plots all curves in the same panel; "rows" splits each curve into a separate row with features in columns; "cols" splits each curve into a separate column with features in rows. Use fixed_scale_curves = TRUE to fix the y-axis upper limit to the maximum value across all curves for each feature.

  • r2_vstep — adjusts the vertical spacing between R² labels of different curves when curve_layout = "overlay".

mrmhub 0.9.1

mrmhub 0.2.5

mrmhub 0.2.4

mrmhub 0.2.3

mrmhub 0.2.2

mrmhub 0.2.1

mrmhub 0.2.0

Fully revised and extended functions, data classes, unit tests, and documentation.

mrmhub 0.1.0

Initial version