Map PAVE pave_seed (m/z) to vertex group labels for network plots

In PAVE exports, pave_seed is the metabolite seed m/z, not a feature ID. Each pave_seed is matched to the closest m/z in pave.trace.merged (within ppm) to obtain a seed feature_id group label. PAVE metabolites without pave_seed are their own seed (feature_id). Other nodes without pave_seed are labeled "noise" (plot as grey). When no merged feature is within ppm, features cluster by the seed m/z string.

Usage

map_pave_seed_vertex_group(
  feature_id,
  pave_seed,
  pave.trace.merged,
  type_pave = NULL,
  ppm = 10,
  mz_col = "mz_pave",
  id_col = "feature_id"
)

Arguments

feature_id

Feature IDs for nodes in the demo subgraph.

pave_seed

PAVE seed m/z per node (may be NA).

pave.trace.merged

Merged PAVE/TRACE table used for m/z lookup.

type_pave

PAVE annotation type per node (used to detect metabolites).

ppm

Maximum ppm error when matching pave_seed to a feature m/z.

mz_col

Column in pave.trace.merged with m/z values.

id_col

Column in pave.trace.merged with feature IDs.

Value

Named character vector of group labels keyed by feature_id. Value "noise" means no PAVE seed.