Skip to main content
COMETS Analytics

Welcome to COMETS Analytics!

COMETS Analytics supports and streamlines consortium-based analyses of metabolomics data. The software is continuously being developed and maintained by Ewy Mathé (Division of Preclinical Innovation, National Center for Advancing Translational Sciences), Steve Moore (Division of Cancer Epidemiology and Genetics, National Cancer Institute), and Marinella Temprosa (Dept. of Biostatistics and Bioinformatics, Milken Institute School of Public Health, George Washington University), with web interface support from NCI's CBIIT and R package development support from IMS. Constructive feedback is provided by the COMETS Data Infrastructure Working Group and other working groups.

COMETS Analytics was designed to simplify meta-analysis at the consortia level. Users prepare data input, and then the software takes care of checking the data integrity, performs data analyses securely, and aggregates results in a standardized format. Further details on the vision for implementing the software and the current features available can be found here.

Go to the Correlation tab to get started, or to the About tab to learn more!

Select Input Files
Select Template

Conduct cohort-specific correlation analyses

Select Input Files
Uploading....please wait
Specify Method Of Analyses Specify Method Of Analyses
Enter Email

Input fields

Specify the name of your cohort. If your cohort is not listed, please notify the harmonization group.

Integrity check

The tab will display the results of the harmonization and integrity check results.


The table of correlations and metabolite information is displayed with an option to download the results in CSV format. For each exposure, the correlation, n and p value are displayed. Please download the file for each model and send to data harmonization group.


The heatmap displays the metabolites in rows and exposures in columns using partial (if adjustment variables are specified) spearman correlations. From this tab, you can download the heatmap as a png file, pan, zoom and get correlation coefficients. You can also sort the results using the dropdown menu.

Cluster and heatmap

The cluster and heatmap is available for use when there are 2 or more exposures and 2 or more outcomes are specified.


The software can be run two different ways:

  1. A standalone R package “R-cometsAnalytics” that encapsulates our core algorithms and functionality, allowing the software to be run locally. The GitHub repository for the development of COMETS Analytics is publicly available at under a GPL-3 license.
  2. A web-based app ( developed as a user friendly interface to the R package using HTML5. This app operates on secure cloud-based servers that delete data after analyses.

All underlying statistical analyses and data processing use the R-cometsAnalytics R package so that using the R package or the app will produce the same results.

Current Version Functionality

Current Version is 1.6: Released on 10/14/2019 with analytic module for unadjusted and partial correlation analyses. Complete details of the version history are documented in the GitHub repository:

Upcoming Version 2.0: Analytic module with generalized linear models is in testing, expected release in 2021.

Previous releases can be found here:


A companion tutorial can be found at A presentation of the software, its implementation and vision can be found here.

For questions or help on COMETS Analytics app or R package, please send an e-mail to

For help on metabolite harmonization, contact Steve Moore.


We thank the National Cancer Institute (NCI) for supporting the development and expansion of COMETS Analytics, the NCI Center for Biomedical Informatics and Information Technology team for developing the app, the Information Management Services team for further developing the R package, and our users for providing feedback so we can continuously ameliorate the software.

Web application Development Team: Kailing Chen, Ewy Mathé, Steve Moore, Brian Park, and Ella Temprosa.

Metabolite Harmonization Team: Dave Ruggieri and Steve Moore.

A special thanks as well to the Broader COMETS Data Infrastructure Group.


Please site the following when using COMETS Analytics:

Temprosa M, Moore SC, Zanetti KA, Appel N, Ruggieri D, Mazzilli KM, Chen KL, Kelly RS, Lasky-Su JA, Loftfield E, McClain K, Park B, Trijsburg L, Zeleznik OA, Mathé EA. COMETS Analytics: An online tool for analyzing and meta-analyzing metabolomics data in large research consortia. Am J Epidemiol. 2021 Apr 22:kwab120. doi: 10.1093/aje/kwab120. Epub ahead of print. PMID: 33889934.

Loading....Please Wait