SessionRegFileExprTxt

Home | Expression Sessions | Files

Expression matrix [expr.expr.txt or GEO.expr.txt]

Each row of the expression matrix has the expression values for a single gene (or feature) across all of the samples. Here are some important facts about this file:

• the file is TAB delimited
• the first row is the header, and it is only the sample ids, with no column name in the header for the gene name. In other words, the first word of the first row is the first sample id, not something else like "gene_name", "id", etc.
• the sample ids in the header must be the same order and count as the row names (sample ids) of the samples.csv file
• the sample ids in the header should be follow standard R variable naming convention (i.e. must start with a letter and can only be a combination of letters, digits, ., and _)
• the first field of each row is the gene (or feature) name
• none of the gene (feature) names can be repeated -- even if the case is different, they must not be equal.
• the data is loaded from this file, so I often like to decrease the number of significant digits to make uploading and accessing of the data faster.

Note that many of these rules can be skipped at first with the expression matrix file that you upload, but then must be modified before the expression matrix file is loaded to the SQL portion of the app.

Here is a the upper right corner of an example file: