Skip to content

Concepts

The Challenge: Isolated Metabolomics Studies

Metabolomics studies generate rich data about small molecules in biological samples. However, most studies exist as isolated datasets:

  • Each study uses different experimental designs
  • Sample groups and conditions vary across studies
  • Raw data formats and processing methods differ
  • Results are difficult to compare directly

This limits the potential for reuse and discovery across studies.

The Solution: Differential Profiles

integMET addresses this challenge by introducing the Differential Profile (DiffProf) as the fundamental unit of data.

What is a Differential Profile?

A Differential Profile captures how metabolites change between two biological conditions.

For example: - Disease vs. Control - Before treatment vs. After treatment - Mutant vs. Wild type - Stressed vs. Normal

Each DiffProf contains: - Two comparison groups (Group A vs. Group B) - Metabolite fold-changes between the groups - Statistical significance (p-values) for each change - Direction of change (up or down)

From Raw Data to Comparable Profiles

Traditional metabolomics produces a data matrix: rows of metabolites, columns of samples, cells containing abundance values.

integMET converts this into DiffProfs:

Traditional Data Matrix       →      Differential Profile

  Sample1  Sample2  ...              Group A vs Group B
  Met1: 100    150                   Met1: ratio=1.5, ↑
  Met2: 200     80                   Met2: ratio=0.4, ↓
  ...                                ...

By standardizing data into DiffProfs, integMET makes different studies comparable.

Why DiffProfs Enable Integration

1. Universal Comparison Unit

Every biological comparison — regardless of species, tissue, or experimental platform — becomes a DiffProf. This creates a common language for metabolomics data.

2. Focus on Biological Contrast

A DiffProf represents a biological event or contrast, not just raw measurements. This captures the biologically meaningful signal: what changes when conditions differ.

3. Cross-Study Discovery

With thousands of DiffProfs in a standardized format, you can:

  • Find studies with similar metabolic signatures
  • Discover unexpected connections between different biological phenomena
  • Identify metabolites that consistently change across related conditions

Key Terminology

Term Definition
Study A metabolomics experiment from a public repository (e.g., MetaboLights)
Differential Profile (DiffProf) A standardized record of metabolite changes between two conditions
Metabolite A small molecule tracked across studies and DiffProfs
iDMET A similarity metric that quantifies how similar two DiffProfs are based on shared metabolite changes

Next Steps

  • Learn about the Data Model to understand how Studies, DiffProfs, and Metabolites relate
  • Explore the Tutorial to see how DiffProf similarity networks work