UCLA Gene Expression Tool (UGET)

Gene-Gene Correlation in Celsius

This tool gives the correlation matrix for two given Affymetrix probeset/gene lists across thousands of samples within a given platform in Celsius.
Please paste a list of Affymetrix gene identifiers in List No. 1. The other list can be a list of gene identifiers OR will be retrieved based on chromosome alignment information.

Other tools: (1) HDN: The Human Disease Network gene query tool [LINK]
To cite the UGET tool, please use the following reference:
Day A, Carlson MR, Dong J, O'Connor BD, Nelson SF. (2007) Celsius: a community resource for Affymetrix microarray data. Genome Biol. 2007;8(6):R112   PMID: 17570842 
Day A, Dong J, Funari VA, Harry B, Strom SP, et al. (2009) Disease Gene Characterization through Large-Scale Co-Expression Analysis. PLoS ONE 4(12): e8491. doi:10.1371/journal.pone.0008491

Select a platform:

Select a gene identifier type:

Output the correlation matrix as Excel file?: Yes No  (Selection of 'Yes' may take longer to return the results. Select 'No' for a quick view.)

List No.1 becomes the Columns, and List No. 2 becomes the Rows. The quick view gives the average across items in List No. 1 for each item in List No. 2.

List No. 1 List No. 2
Option 1: Paste one or more Gene Symbols. Option 2: Select a chromosome and specify the range: (This gives all genes in this range on this chromosome.)

Start                      (numeric digits only)

End                       (numeric digits only)



  1. List no. 1 can not be empty!
  2. If List No. 2 is empty AND Option 1 is selected, then List No. 1 will be copied to generate a symmetric correlation matrix.
  3. One row for each identifier (probeset ID or gene symbol). Duplicate rows will be automatically removed.
  4. Remove any blank space before and after each identifier. This may cause an identifier unrecognizable, which will be removed from the output.
  5. All unrecognized identifiers will be listed. Please double check them for possible typos or trailing spaces.
  6. Any type of input identifier will be converted to matching g probeset ID in the output table.
  7. The output table will always put List No. 2 as rows, and List No. 1 as columns.
  8. To retrieve correlation coefficients with all available probesets, set one field as "ALL". (The value in field #2 is set to "ALL" in default. Field #1 can NOT be empty.)
  9. Due to the data retrieving time & load reasons, the length of the shorter list must not exceed 500. Therefore ("ALL", "") and ("ALL","ALL") etc are not allowed.
  10. For Option 2, the length of List No. 1 must not exceed 500.
  11. The output 2-way table is an optional downloadable tab-delimited file with .xls extension.
  12. The top 300 mean correlation coefficients across items in List No. 1 for each item in List No. 2 will be listed in descending order.

Examples of common tasks:

  1. To obtain correlations of a list with all available probesets:  Field #1 = Your list; Field #2 = "ALL" (default value)
  2. To obtain correlations of a list with itself (symmetric correlation matrix):  Field #1 = Your list; Field #2 = "" (empty)
  3. ("ALL", "") and ("ALL","ALL") etc are not allowed.
(Last updated: 03/27/2009 12:02 AM)