Skip to contents

Welcome to eCOMET: A Tool for Mass Spectrometry–Based Ecological Metabolomics

eCOMET is an R package for processing and analyzing mass spectrometry–based metabolomics data in ecological and evolutionary contexts. It establishes a standardized pipeline that integrates MS1 feature–abundance tables with MS2 spectral similarity to generate common data products, including principal component analyses (PCA), chemical dendrograms where tips represent compounds, principal coordinates analyses (PCoA), and differential accumulation analyses. Unlike existing metabolomics toolkits that focus primarily on biomedical or cheminformatics applications, ecomet is designed explicitly for ecometabolomics. It emphasizes workflows that link metabolomic variation to ecological data, making it easier to move from raw mass spectrometry files to reproducible, comparative analyses of plant chemical diversity.

Key Features

  • Import and normalize MS1 feature–abundance tables
  • Integrate MS2 spectral similarity for chemical diversity analysis
  • Perform differential accumulation analysis to find DAMs (Differentially accumulated metabolites)
  • Visualize results with PCA, PLS-DA, PCoA, NMDS, volcano plots, heatmaps, and more
  • Quantify chemical diversity using alpha and beta diversity metrics, utilizing spectral similarity based on modified cosine score, MS2DeepScore, and DreaMS
  • Chemical class enrichment analysis using CANOPUS annotation
  • Screening molecular features related to sample metadata (e.g., temperature, height, herbivory)

Screenshot 2025-07-24 at 6 59 47 PM

Installation

Install the development version from GitHub:

# install.packages("pak")
pak::pak("phytoecia/eCOMET")
library(ecomet)

Preparing data for analysis

We recommend reading Articles-Intro before starting, to understand how the raw data from mass spectrometers can be prepared for analyses.
We currently support data processed by MZMine (version 4 or later).

Tutorials

eCOMET can be used for chemical ecological studies from diverse spectra of interests. We prepared two case studies: 1. Treatment-based_study 2. Observational_field_study

For more details, see the documentation and vignettes.
If you have questions or suggestions, visit the GitHub repository, where we are actively developing.