Functional Level Metabolic Annotation

---
name: functional-level-metabolic-annotation
description: Use when you have reconstructed the structural topology of two metabolic
  networks from KEGG and need to assign functional meaning (pathway annotations, metabolic
  functions) to each node or reaction so that you can compare the two organisms at
  the functional level rather than only at the topological.
license: CC-BY-4.0
metadata:
  edam_operation: http://edamontology.org/operation_3662
  edam_topics:
  - http://edamontology.org/topic_0602
  - http://edamontology.org/topic_0092
  - http://edamontology.org/topic_3407
  tools:
  - MetNet
  - Java
  - KEGG
  license_tier: open
  provenance_tier: literature
derived_from:
- doi: 10.1371/journal.pone.0246962
  title: MetNet
evidence_spans:
- MetNet is a Java tool that makes it possible to automatically reconstruct the metabolic
  network of two organisms selected in KEGG
- MetNet is a Java tool that makes it possible to automatically reconstruct the metabolic
  network
- MetNet is a Java tool
claims: []
provenance:
  collection: https://w3id.org/holobiomicslab/asb-skill/collection/metabolomics/v2
  assembled_by: scripts/collect_metabolomics_collection.py
  sources:
  - build: coll_metnet_cq
    doi: 10.1371/journal.pone.0246962
    title: MetNet
  dedup_kept_from: coll_metnet_cq
schema_version: 0.2.0
attribution:
  generator: AgenticScienceBuilder
  original_doi: 10.1371/journal.pone.0246962
  all_source_dois:
  - 10.1371/journal.pone.0246962
  zenodo_doi: 10.5281/zenodo.20794027
  curators: []
  promoter: Louis-Félix Nothias
  sponsor: CNRS & Université Côte d'Azur
---

# functional-level-metabolic-annotation

## Summary

Annotate metabolic networks with functional-level representations that map metabolic functions and pathway annotations to network nodes, enabling quantitative comparison of pathway-level biology between organisms. This skill complements structural topology analysis by capturing what metabolic processes occur rather than just how they are connected.

## When to use

You have reconstructed the structural topology of two metabolic networks from KEGG and need to assign functional meaning (pathway annotations, metabolic functions) to each node or reaction so that you can compare the two organisms at the functional level rather than only at the topological level. Use this skill when your research question requires understanding whether two organisms perform similar metabolic functions, even if their network topologies differ.

## When NOT to use

- You only need to compare network topologies (connectivity patterns, degree distributions, shortest paths) — use structural-level comparison instead.
- Your input organisms are not available in KEGG or their KEGG records lack pathway annotations.
- You have pre-computed functional annotations from a different source and only need to map them to an existing topology — use annotation mapping instead of re-annotating from KEGG.

## Inputs

- Structural metabolic network topology (nodes and edges from KEGG for each organism)
- KEGG organism identifiers (e.g., 'hsa' for Homo sapiens, 'ptr' for Pan troglodytes)
- KEGG pathway and reaction database records
- Metabolic function definitions and pathway annotations from KEGG

## Outputs

- Functional-level network representation mapping metabolic functions to pathways
- Pathway annotation assignments for each organism
- Functional-level similarity indexes between the two organisms
- Structured output file containing both sets of functional similarity scores

## How to apply

Load the structural metabolic network topology for each organism (generated by KEGG data retrieval and MetNet reconstruction). Extract functional features by querying KEGG pathway annotations and metabolic function assignments for each reaction or pathway in the network. Map these functional annotations (enzyme classifications, pathway names, metabolic roles) to the corresponding nodes or edges in the structural network. Execute MetNet's functional-level representation module to generate a mapping layer that associates each pathway with its constituent metabolic functions. Compute functional-level similarity indexes by comparing pathway annotations and metabolic function assignments between the two organisms. The rationale is that functional annotation reveals biological meaning independent of network topology; two organisms may have different connection patterns but similar metabolic capabilities.

## Related tools

- **MetNet** (Executes functional-level network representation generation and computes functional-level similarity indexes between organism metabolic networks) — github.com/simeoni-biolab/MetNet
- **KEGG** (Source database for metabolic pathways, reaction data, and functional annotations used to assign metabolic functions to network nodes)
- **Java** (Runtime environment for MetNet tool execution)

## Examples

```
java -jar MetNet.jar hsa ptr set
```

## Evaluation signals

- Functional-level similarity indexes are computed and reported for both organisms — verify that the output file contains non-empty functional similarity scores.
- Each pathway in the structural network has been assigned metabolic function annotations from KEGG — verify that the annotation layer covers all major pathways with non-null function assignments.
- Functional similarity scores differ from structural similarity scores — if they are identical, the functional annotation layer may not have added information.
- Aggregated similarity report compares pathway annotations and metabolic function assignments between the two organisms — verify that the output distinguishes shared vs. organism-specific metabolic functions.
- Metabolic functions are correctly mapped to KEGG pathway and reaction identifiers — spot-check that enzyme classifications and pathway names match KEGG database records.

## Limitations

- Functional annotation depends entirely on KEGG pathway and annotation quality — organisms or pathways with sparse KEGG records will have incomplete functional representations.
- The two-level representation (structural + functional) is designed for pairwise organism comparison; comparison of more than two organisms requires extension of the similarity index methods.
- Functional similarity indexes are sensitive to the choice of pathway list (configured in pathwayList.txt) — different subsets of pathways may yield different comparison results.
- MetNet's functional-level representation assumes that KEGG annotations are complete and consistent; metabolic functions outside KEGG's pathway database will not be captured.

## Evidence

- [intro] Extract functional features by comparing metabolic function assignments: "Extract functional features (pathway annotations, metabolic functions) from each organism's functional representation."
- [intro] A two-level representation with functional level for metabolic functions: "a two-level representation of metabolic networks: a structural level representing the metabolic network topology and a functional level representing the metabolic functions of each pathway"
- [intro] Compute functional similarity by comparing metabolic functions: "Compute functional-level similarity indexes by comparing metabolic function assignments and pathway annotations between the two organisms."
- [readme] MetNet generates functional-level representation mapping: "their metabolic data are retrieved from KEGG and the corresponding networks of metabolic functions are built"
- [readme] Similarity indexes support comparison at both levels: "The approach is supported by similarity indexes for the comparisons at both levels."