Skip to content

The MapMan pathways database

MapMan is a plant-oriented functional annotation. Instead of curated reaction networks it offers a numbered ontology of bins — functional categories — and a set of hand-drawn diagrams that place those bins on a picture of plant metabolism. In PaintOmics it is one of the optional pathway sources, offered for a small number of plant organisms.

The application describes it on the Step 2 mapping screen as oriented towards plant species, in combination with GoMapMan, it provides additional pathways as well as an improved and more consolidated annotation for the model species Arabidopsis, and several crop species (potato, tomato, rice).

Which organisms actually have it

Four, in the shipped configuration:

Organism Code
Arabidopsis thaliana ath
Tomato (Solanum lycopersicum) sly
Potato (Solanum tuberosum) sot
Rice (Oryza sativa) osa

Two warnings about that list.

Rice has two KEGG codes and only one of them carries MapMan. osa has it; dosa, the RAP-DB-keyed rice entry, does not, because KEGG publishes no NCBI-gene-id conversion for dosa and the gene-to-bin cross-link has nothing to join through. If you work on rice and want MapMan diagrams, choose the organism whose code is osa.

Sugar beet (bvu) has a MapMan build but no MapMan identifier mapping in the shipped configuration, so its MapMan checkbox stays disabled and only its KEGG pathways are analysed.

As with the other optional databases, whether your server has MapMan installed for the organism you picked is what the Step 1 checkbox says: tickable when it is there, disabled and labelled not installed when it is not.

Bins, and what a MapMan box actually holds

A MapMan bin is a dotted number: 1 is a top-level function, 1.3 a subcategory of it, 1.3.4 a subcategory of that. A gene-to-bin file assigns each gene one or more bins, and PaintOmics stores those gene names as their own identifier type, mapman_gene_id.

A diagram does not place genes. It places bins: each area of the picture names one bin, and PaintOmics expands that bin to every gene mapped to it or to any bin nested underneath it — an area labelled 18.4 collects the genes of 18.4, 18.4.1, 18.4.1.2 and so on.

The consequence is worth understanding before you read a painted MapMan diagram: one box on a MapMan diagram is a bin, not a gene. Every matched gene in that bin is painted into the same box, so a box can stand for hundreds of genes where a KEGG box stands for the handful of gene products drawn at that spot. Click it and the feature detail panel lists what is in it.

The diagrams

MapMan pathways in PaintOmics are its diagrams, and they are named rather than numbered — Glycolysis-TCA, Core metabolism overview, Metabolites. There are about seventy of them, assembled from two sources:

  • GoMapMan's own PaintOmics export ships twenty, overwhelmingly Secondary Metabolism and Hormones;
  • the installer adds fifty more from the MapManStore archive — the general maps a MapMan user actually reaches for: metabolism overview, glycolysis, TCA, photosynthesis, transcription, and the Metabolites compound map.

All of them are MapMan 3.6-era diagrams, matching the bin numbering GoMapMan's gene mappings use. X4-era diagrams are deliberately excluded: they renumber the ontology, and would place the wrong genes without reporting an error.

The extra fifty are fetched all-or-nothing. That is deliberate — the number of pathways is a denominator in the enrichment test, so an install that fetched sixty-three diagrams today and sixty-seven tomorrow would silently change p-values.

Because the pathway identifier is a diagram name rather than an accession, the external-link button in the enrichment table searches GoMapMan for that name rather than opening a record.

Metabolites

MapMan ships a metabolite mapping of its own — compound symbols against bin codes — which PaintOmics loads alongside the KEGG compound names. A metabolite name in your file can therefore resolve to a MapMan bin and be painted on a MapMan diagram, the Metabolites map in particular.

Where KEGG publishes an NCBI-gene-id conversion for the species, the installer links each MapMan gene to the matching KEGG gene ids, so one identifier from your file can be painted on both a KEGG map and a MapMan diagram. Where that conversion does not exist, MapMan genes still install and still map — they simply stand on their own, unlinked to the KEGG side.

Classification

MapMan diagrams carry two classification levels from MapMan's own classification file: a primary category and a secondary one, for example Metabolism then Amino Acids. Step 3's category pie and Filter by category tree use those two levels, as they do for the other databases. A diagram the classification file does not name is filed as Not classified / Unclassified.

For how the four databases differ, see the comparison table on the KEGG page.