Submitting Data to ArrayExpress

ArrayExpress is a public repository for transcriptomics and related data at the EBI. You can submit data to ArrayExpress in advance of publication and we will keep it private and provide a reviewer login for you. This help page explains how to submit to ArrayExpress. There is general information about ArrayExpress here:
ArrayExpress help page >>

In this help page:

What you can submit

ArrayExpress accepts submission of all types of microarray data, microarray designs, and supporting protocols and sample information. The types of microarray data we support include gene expression, ChIP-chip, comparative genomic hybridization (CGH) and protein arrays. Both Affymetrix and 2-channel data is supported.

How to submit experiments

Submit data to ArrayExpress using one of our two submission routes:

MIAMExpress web based submission tool

  • this tool requires you to fill in a series of forms describing your samples and protocols, and then upload data files for each of your hybridizations
  • a batch upload tool is available which allows you to describe all your samples in a tabular format
  • best suited to small to medium sized experiments (up to 50 hybridizations if using the batch upload tool)
  • 1 raw data file is required for each hybridization
  • normalized data (either 1 file per hybridization or 1 summary file containing data from all hybs) can be uploaded in addition to raw data and we strongly encourage this
  • MIAMExpress submission help page >>

Tab2MAGE/MAGE-TAB spreadsheet submission systems

  • these submission routes require you to fill in a spreadsheet describing your samples and protocols and then upload this to us along with your data files
  • the submission systems can create a template experiment description spreadsheet to get you started
  • suitable for any size of experiment
  • any number of raw data files can be provided for each hybridization
  • multiple normalized or summary data files be provided
  • Tab2MAGE/MAGE-TAB must be used to submit Illumina, Nimblegen or 2-channel Imagene data. MAGE-TAB must be used for UHTS (Solexa and 454) data
  • Tab2MAGE/MAGE-TAB are strongly recommended for large experiments (50+ hybridizations) or if you have large amounts of sample annotation already in a spreadsheet (e.g. a table of clinical history information)
  • The differences between Tab2MAGE and MAGE-TAB are outlined on the Tab2MAGE vs MAGE-TAB comparison page. If you are already using Tab2MAGE for your submissions you do not need to change to MAGE-TAB.
  • spreadsheet submission help page >>

How to submit array designs

If you used a custom or self-spotted array design you will need to submit the array layout to us using MIAMExpress. There is more information on submitting array designs here:
Submitting array designs >>

Accession number assignment and privacy

After you have submitted your experiment or array design to ArrayExpress:

  • The submission will be reviewed by the curation team.
  • If all the necessary information has been provided we will assign an ArrayExpress accession number to the submission and begin loading it into ArrayExpress. This process can take several days and we will contact you by email if we have any questions.
  • After we have loaded your submission into ArrayExpress we will provide you with the accession number and a reviewer login account that can be used to view the data in ArrayExpress.
  • Many journals require accession numbers for microarray data before acceptance of a paper for publication. Please submit your data to us well in advance of this.
  • If you urgently need the accession number or reviewer login details for a completed submission then email us at and we will answer as soon as possible.

Data release policy:

  • All information you enter into the MIAMExpress or Tab2MAGE/MAGE-TAB submission tools is private and can only be viewed by you and the curation team.
  • We ask you to provide an estimated public release date when making your submission. After your submission has been loaded into ArrayExpress it will remain private until your requested release date.
  • If no release date is supplied we will set it to one year from the date the submission was loaded.
  • The submitter of the experiment or array design will receive a reminder email 30 days before the submission is made publicly available to allow you to postpone the release date.
  • Reminder emails will be sent to the contact email address provided at the time of submission - it is your responsibility to tell us if the contact email address changes (email the new address to ).
  • Email us at at any time to request a change to a public release date.
  • If we find that you have published a paper referencing data that has been deposited in ArrayExpress then we will make the data publicly available.

ArrayExpress submission, curation and loading process

MIAME compliance

MIAME describes the Minimum Information About a Microarray Experiment that is needed to enable the interpretation of the results of the experiment unambiguously and potentially to reproduce the experiment.

All submissions made using the MIAMExpress or Tab2MAGE/MAGE-TAB systems are reviewed by the curation team and any MIAME required information that has not been provided will be requested. If the requested information is not supplied then the submission will be processed without it. After loading into ArrayExpress all experiments are given an automatically generated MIAME compliance score between 0 and 5, where 0 means that none of the MIAME requirements were met and 5 means that all requirements were met. This score allows reviewers and other users to assess how much MIAME required information was provided for an experiment.

Details about what information is required for MIAME compliance can be found here:
MIAME homepage and checklist >>

Pipeline submissions

If you are a software developer and you would like to submit large amounts of data to ArrayExpress from your own microarray database on an on-going basis then you can set up a MAGE-TAB submission pipeline. The content, accuracy and MIAME compliance of pipeline submissions are the responsibility of the organization generating the MAGE-TAB files. Read our documentation before attempting to set up a MAGE-TAB pipeline:
Pipeline submission help page >>