Build a reusable indexed metabolite annotation search space

Expected adduct m/z values are generated once, sorted, and queried by binary interval search. This is the one-dimensional equivalent of an interval tree and avoids scanning every database row for every observed peak.

Usage

BuildMZAnnotationIndex(
  db,
  polarity = NULL,
  adducts = NULL,
  rules = NULL,
  maldi_matrix = NULL,
  collapse_isomers = TRUE
)

Arguments

db

Metabolite database. Both the legacy SpaMTP five-column/wide database and the RaMP chem_props schema are supported.

polarity

"positive", "negative", or "neutral". When NULL, use the matrix-profile default, or positive mode when no profile is given.

adducts

Optional character vector of adduct names or bracketed notations. When NULL, use the complete automatically selected rule space; this is a filter, not a compulsory input.

rules

Optional custom rule table. See AdductRules().

maldi_matrix

Optional MALDI matrix or derivatization reagent profile. When supplied and rules is NULL, MALDIMatrixRules() automatically selects standard and validated matrix-specific rules. adducts remains an optional filter on that selected rule space.

collapse_isomers

Collapse records sharing formula, exact mass, and proton bound before indexing.

Value

An object of class spamtp_mz_index.