Skip to contents

Import MiSoSoup YAML Output

Source: R/importMisosoup.R
importMisosoup.Rd

Imports raw MiSoSoup YAML output into a ConsortiumMetabolismSet object. MiSoSoup YAMLs describe many consortia per file (one per substrate / second-level-key / solution triple), so a single call always produces a CMS regardless of whether the input is a single file, a directory of files, or a pre-loaded nested list.

Usage

importMisosoup(
  data,
  name = NULL,
  normalize_ids = TRUE,
  biomassPattern = "(?i)^.*?(?:^|_)(?:biomass|growth)_",
  verbose = TRUE
)

Arguments

data

Either a file path to a single MiSoSoup YAML, a directory path containing multiple YAML files, or a pre-loaded nested list from yaml::read_yaml().

name

Name for the returned ConsortiumMetabolismSet. If NULL (default), the name is derived from the input filename (for a file) or directory basename (for a directory). Required when data is a pre-loaded list.

normalize_ids

If TRUE (default), normalize metabolite IDs via .normalizeBiggIds() (strip R_EX_ prefix, _e suffix) and decode COBRApy __NN__ escape sequences via .decodeBiggEscapes(). If FALSE, keep metabolite IDs in their raw form.

biomassPattern

Perl-compatible regex identifying biomass reaction rows. The pattern must match and consume the entire prefix from the start of the reaction ID up through the separator before the species ID, because it is used for both detection and species extraction (via sub()). Defaults to "(?i)^.*?(?:^|_)(?:biomass|growth)_", which covers the three observed MiSoSoup variants (Growth_<sp>, R_Biomass_<sp>, R_R_BIOMASS_<sp>) case-insensitively and correctly rejects the community_growth summary row. Override this if your MiSoSoup output uses a non-standard biomass reaction name, e.g. biomassPattern = "(?i)^.*?MYBIO_".

verbose

If TRUE (default), emit progress messages via cli::cli_inform() and show a progress bar when importing a directory.

Value

A ConsortiumMetabolismSet object containing one ConsortiumMetabolism per viable consortium found in the input. Zero-growth solutions are silently skipped.

Details

The function auto-detects format differences between MiSoSoup runs: the biomass-reaction name (via a configurable regex that defaults to matching Growth_<sp>, R_Biomass_<sp>, and R_R_BIOMASS_<sp> case-insensitively — MiSoSoup derives this name from the underlying model, so it varies), COBRApy-style __NN__ escape sequences in reaction IDs (e.g. __40__ for (), and the second-level key of each substrate. That second-level key is either min — indicating a CMSC (complete minimal supplying community: no viable focal strain because no single member grows alone) — or a focal-strain ID, indicating an MSC (minimal supplying community for that strain).

Per-consortium metadata set on each returned CM:

  • metadata(cm)$media — media-level exchange fluxes (reactions without a species suffix), preserving the original bounds.

  • metadata(cm)$communityGrowth — the scalar community_growth summary row from the MiSoSoup solution, or NA_real_ if absent.

  • metadata(cm)$misosoupMode — "CMSC" or "MSC".

  • metadata(cm)$focalStrain — focal-strain ID (character) for MSC consortia, NA_character_ for CMSC.

See also

importSmetana() for the sibling SMETANA import.

Examples

# Build a minimal in-memory MiSoSoup-shaped list and import it. In
# practice you would point at a real YAML file or a directory of
# YAMLs from a MiSoSoup run.
raw <- list(
    glucose = list(
        min = list(
            list(
                community = list(y_sp1 = 1, y_sp2 = 1),
                solution = list(
                    R_Biomass_sp1 = 0.4,
                    R_Biomass_sp2 = 0.3,
                    R_EX_glc__D_e_sp1_i = -10,
                    R_EX_ac_e_sp1_i = 5,
                    R_EX_ac_e_sp2_i = -4,
                    R_EX_co2_e_sp2_i = 3,
                    community_growth = 0.7
                )
            )
        )
    )
)
cms <- importMisosoup(raw, name = "demo", verbose = FALSE)
#> 
#> ── Creating CMS "demo" ─────────────────────────────────────────────────────────
#> ℹ Validating 1 <ConsortiumMetabolism> object
#> ✔ Validating 1 <ConsortiumMetabolism> object [11ms]
#> 
#> ℹ Collecting metabolites from 1 consortia
#> ✔ Collecting metabolites from 1 consortia [27ms]
#> 
#> ℹ Re-indexing 3 unique metabolites
#> ✔ Re-indexing 3 unique metabolites [26ms]
#> 
#> ℹ Expanding 1 binary matrices to 3-dimensional space
#> ✔ Expanding 1 binary matrices to 3-dimensional space [22ms]
#> 
#> ℹ Computing 3 x 3 levels matrix
#> ✔ Computing 3 x 3 levels matrix [31ms]
#> 
#> ℹ Computing pairwise overlap (0 pairs via crossprod)
#> ✔ Computing pairwise overlap (0 pairs via crossprod) [20ms]
#> 
#> ℹ Assembling pathway data from 1 consortia
#> ✔ Assembling pathway data from 1 consortia [28ms]
#> 
#> ℹ Collecting 1 consortium graphs
#> CMS "demo" created: 1 consortia, 3 metabolites (0.2s)
#> ✔ Collecting 1 consortium graphs [76ms]
#> 
cms
#> 
#> ── ConsortiumMetabolismSet 
#> Name: "demo"
#> 1 consortia, 2 species, 3 metabolites.
#> Community size (species): min 2, mean 2, max 2.
#> Community size (metabolites): min 3, mean 3, max 3.

# Real-world usage (commented; needs an actual MiSoSoup YAML on disk):
# cms <- importMisosoup("path/to/misosoup.yaml")
# cms <- importMisosoup("path/to/misosoup_dir/", name = "experiment1")