Skip to contents

Changelog

Source: NEWS.md

ramen (development version)

Breaking changes

  • EffectiveConsumption and EffectiveProduction assays now store the flux-corrected effective flux F⋅2H(p)F \cdot 2^{H(p)} (matching equation 2 of the underlying thesis), not the bare Hill-1 perplexity 2H(p)2^{H(p)}. Same units as the Consumption / Production assays. Pre-Bioconductor release, no users yet, so no deprecation cycle.
  • Two new assays expose the previous quantity under explicit names: nEffectiveSpeciesConsumption and nEffectiveSpeciesProduction carry the Hill-1 effective number of contributing species (unitless, ∈[1,S]\in [1, S]), mirroring the existing nSpecies count. Algebraic identity: EffectiveConsumption = Consumption * nEffectiveSpeciesConsumption (modulo two-decimal rounding).
  • consortia() no longer has a ConsortiumMetabolism method. The plural noun applies only to containers of consortia, so the accessor is now scoped to ConsortiumMetabolismSet (returns the list of constituent CMs). For the underlying edge list of a single ConsortiumMetabolism, use as.data.frame(cm). consortia(cma) continues to raise an error – by design, ConsortiumMetabolismAlignment is a result object that records only its inputs’ names rather than retaining copies of them.
  • plotDirectedFlow() parameters renamed to lowerCamelCase throughout, with British-English spelling for colour-related args. The size-bearing arguments (nodeSize, nodeLabelSize, edgeArrowSize) now use ggraph millimetre units rather than igraph cex factors, with defaults adjusted for legibility. This is a hard rename – no aliases for the old names.

New methods

  • as.data.frame() is now defined for ConsortiumMetabolism (returns the edge list with columns met, species, flux) and ConsortiumMetabolismSet (row-binds per-CM edges and prefixes a consortium column). The pre-existing ConsortiumMetabolismAlignment method is unchanged.

Other changes

  • importMisosoup() gains a biomassPattern argument (Perl regex) so users can point it at arbitrary biomass-reaction names. The default matches Growth_<sp>, R_Biomass_<sp>, and R_R_BIOMASS_<sp> case-insensitively (MiSoSoup’s biomass reaction name is model-dependent).
  • importMisosoup() now records CMSC/MSC provenance on each returned CM as metadata(cm)$misosoupMode ("CMSC" or "MSC") and metadata(cm)$focalStrain (character or NA). The community_growth summary row is captured as metadata(cm)$communityGrowth instead of leaking into the media bucket.
  • Internal documentation, comments, and user-facing CLI messages standardised on British English spelling (colour, centre, organise, etc.). External-API identifiers from upstream packages (igraph, ggplot2, ColorBrewer) keep their original spelling. The Language: en-GB DESCRIPTION field will follow in a separate commit.

ramen 0.0.0.9001

New Features

  • Initial development release.
  • Three core S4 classes: ConsortiumMetabolism, ConsortiumMetabolismSet, and ConsortiumMetabolismAlignment, all inheriting from TreeSummarizedExperiment.
  • Import microbial metabolic network data from MiSoSoup YAML format via importMisosoup().
  • Pairwise and multiple alignment of consortium metabolisms using functional overlap scores (FOS), Jaccard, Bray-Curtis, and redundancy overlap metrics.
  • Visualisation methods including heatmaps, network plots, and alignment score plots.
  • Functional group analysis with hierarchical clustering of species by shared metabolic pathways.