8 MS1-based Annotation

8.1 Step 2 — Annotate feature table

Features are annotated against the MS1 database via accurate mass matching. Co-eluting features with similar retention times are grouped into Metabolic Feature Clusters (MFCs) to improve annotation confidence and reduce redundancy.

annotation_table_final <-
  annotate_feature_table(
    feature_table       = feature_table,
    column              = "rp",
    metabolite_database = kegg_compound_ms1, # use downloaded MS1 database
    database_type       = "KEGG",            # "KEGG" or "HMDB"
    ms1_match_ppm       = 15,
    mfc_rt_tol          = 10,
    isotope_number      = 3
  )

Key parameters:

Parameter	Description	Default
column	Chromatographic column type: "rp" (reverse phase) or "hilic"	"rp"
database_type	Annotation database: "KEGG" or "HMDB"	"KEGG"
ms1_match_ppm	Mass accuracy threshold in ppm	15
mfc_rt_tol	Retention time tolerance (seconds) for MFC clustering	10
isotope_number	Maximum isotopes considered per feature	3

Tip: Use "rp" for most lipidomics and general metabolomics experiments. Switch to "hilic" for polar metabolite profiling.

8.2 Step 3 — Remove redundant annotations

After initial annotation, redundant entries arising from isotopes, adducts, and in-source fragments are removed. The cleaned annotation table is then processed to produce the ranking table and annotation table used in enrichment analysis.

annotation_table_final2 <-
  featuremsea::remove_redundancy(
    annotation_table = annotation_table_final
  )

results <- process_annotation_table(
  annotation_table_final2 = annotation_table_final2,
  database_type            = "KEGG"   # "KEGG" or "HMDB"
)

ranking_table    <- results$ranking_table
annotation_table <- results$original_score_annotation

The two output objects feed directly into the enrichment step:

• ranking_table — links each feature to its ranking statistic used for enrichment scoring.
• annotation_table — links features to their candidate metabolite annotations.