Understanding ArrayExpress Advanced Results

In this help page:

• Experiment query results
• Array query results
• Protocol query results

This page describes the results produced by querying the ArrayExpress repository for experiments, array designs or protocols using the advanced query interface here: ArrayExpress advanced query interface.

For more information on this and other ways of searching ArrayExpress please see our query help page:
ArrayExpress Query Help page.

Experiment query results

Once an experiment has been found matching the search criteria the information shown in the screenshot below is displayed. If multiple experiments are returned by your query they will all be listed in the same window. The search result for experiments with accession number 'E-MEXP-939' is shown below.

In the top row of each experiment returned is the accession number for the experiment, the submitters name and lab. Below this are an automatically generated description of the experiment, and a free text description provided by the submitter.

The experiment query results include clickable options to see more information:

• Retrieve data
• Experimental protocols
• Providers
• Array design used
• Experiments directory in the FTP/MAGE-ML download/Sample annotation/Experiment design/Data archives
• Bibliographic references
• Samples

Retrieving data

Click on "Retrieve data" to see what data is associated with the experiment. The page is divided into processed (normalized) and measured (raw) data groups. At the top of the page the number of each type of data group is listed. There may be more than one measured and processed data group if quantitation types vary between the files or more than one array design was used. Clicking on the name of the data groups will take you to more information about the data group further down the page.

Some data files are available for download in their native format only e.g. Affymetrix CEL files. In these cases click on the icons in the Data Files column to download the files one by one (files may also be downloaded as a set from the FTP archive). The name of the hybridization/scan and the values for the experimental factors are also shown.

For other data sets you'll see lists of available hybridizations/experimental conditions, quantitation types (measurements) and array annotation as shown here:

Clicking on the names of the hybridizations in the experimental conditions column will retrieve MAGE objects associated with the hybridization.

To export data:

• Select the hybridizations or experimental conditions to export. The list of hybs can be sorted by name or by any of the experimental factors by clicking the column headings.
• Select the quantitation types (measurements) you want. The quanititation types available will depend on what software was used to generate the data. Information about how the data was generated can be found by clicking the protocol and software links above the list of quantitation types. Descriptions of quantitation types can sometimes be found by clicking on the quantitation type name.
• Select the array annotation you want to export with the data. The Reporter identifier or CompositeSequence identifier is a unique identifier for each array element which is provided by the array design submitter. Database entries and oligo sequences are also available if they have been provided by the submitter
• You must select at least 1 hybridization, quantitation type and array annotation.
• Click on the 'Export data' link

The following page will appear while your data matrix is being generated:

Clicking on the "See data matrix" link on this page will retrieve a tab-delimited file containing the data selected for download on the previous screen. The column headers in the data matrices tell you the name of the hybridization, the experimental factor values and the quantitation type of the data in the column. The data matrix can be saved as a tab-delimited text file by using the 'File -> Save As' feature of the browser. Opening the saved text files in a spreadsheet program such as MS Excel will keep the column headers with the correct data columns. The first few lines from a data matrix are shown below:

Note: ArrayExpress does not store images so you can only retrieve tab delimited files, and not the original image data.

Retrieved data may also be uploaded to Expression Profiler, a free tool provided by the EBI for the analysis of expression data. Please see the Expression Profiler web page for further information.

Providers

The list of Providers contains information about the submitters and institutions they are associated with, as in the example below.

Experiments directory in the FTP/MAGE-ML download/Biosamples visualizations

Data files, spreadsheets containing sample and experiment design information in MAGE-TAB format, diagrams of the experiment layout and MAGE-ML files are available for download from our FTP site. Descriptions of the available files can be found here: File downloads from ArrayExpress.

MAGE-ML (Microarray Gene Expression - Markup Language) is the file format used to load experiment and array design information into ArrayExpress. After loading into ArrayExpress we make the original MAGE-ML available for download so that it can be used for loading into other databases. For further information on working with this file format see the MAGE group homepage.

Bibliographic references

The bibliographic reference contains information about any publications associated with the experiment.

Samples

This link will take you to information about each of the original biological materials used in the study (the BioSources). This view provides information on such properties of the starting samples such as species, developmental stage, organism part, gender and cell type. If the samples were treated differently during the experiment then follow the 'Retrieve Data' to see which treatment conditions relate to each hybridization.

Array query results

An example of an array query result displaying a single array is shown below.

The accession number for the array, array name and the provider laboratory are supplied as well as a free text description of the array. An automatically generated summary is also provided.

The array query results include clickable options to see more information:

• Providers - lists contact details for providers of the array design (this could be the array manufacturer or a user of the array)
• Experiments done with this Array - lists all experiments which use this array design. For more information see the experiment query results section.
• Spreadsheet (tab delimited) - opens an Array Design File (ADF) containing the array layout and annotation. More information on the ADF format can be found in the ADF creation help page
• MAGE-ML for download - MAGE-ML (Microarray Gene Expression - Markup Language) is the file format used to load experiment and array design information into ArrayExpress. After loading into ArrayExpress we make the original MAGE-ML available for download so that it can be used for loading into other databases. For further information on working with this file format see the MAGE group homepage.

Protocol query results

Protocols are supplied for the treatments of the samples and experimental procedures including sample growth and treatment, labeling, hybridization, image acquisition and data analysis. Protocols contain free text and sometimes a link to en external web site and a list of protocol parameters.

Any further questions, please see our FAQ.