Sample Sheet

Samples names, and optionally groups, can be defined using a text file with a number of columns, separated by commas (CSV format).    The first column is a well name, used to identify the sample.  The second column is a name for the sample that will appear on charts and in tables.  Other columns define variables, that is, sample groupings.

For instance, a 4x3 experiment might be described as follows
    well, patientId, condition
    A01, pid1, pre-treatment	
    A02, pid2, pre-treatment
    A03, pid3, pre-treatment
    A04, pid4, pre-treatment
    B01, pid5, post-treatment
    B02, pid6, post-treatment
    B03, pid7, post-treatment
    B04, pid8, post-treatment
    C01, pid9, normal
    C02, pid10, normal
    C03, pid11, normal
    C04, pid12, normal
    C05, aneg, NA
    C06, aneg, NA
    C07, apos, NA
    C08, apos, NA
The header rows specifies that there are three columns in the file. The first two columns are always the well location and sample name.  Any number of other columns may be named, which will become named variables.  The variables can be used to pool samples across replicates, or as factors for Anova analysis.   The names should not contains commas or spaces.

In the above example, there are 3 groups, each with 4 patients.  There are 4 other samples which are assay negative and positive controls.   They are assigned group IDs of NA, so that they are not compared to the patient groups during Anova analysis.   (Control samples would presumably always look significantly different to patient samples, and create false significance.)

The table should not be ragged; every row must have the same number of columns.
If the column has no meaningful value in that row (for instance, a negative sample is neither male nor female, neither case nor control patient) the entry should be NA, or empty.   An empty entry is denoted by two commas in immediate succession.  A space between the commas is treated as if it were a name.

The match is based on the well name embedded in the FCS record for that well (A01 etc).

CSV generation

The easiest way to generate a CSV file is with a text editor or spreadsheet of your choice.
If you generate in a spreadsheet, export in CSV format, rather than the native format (e.g. .xlsx or .xls for Excel).

If you have renamed samples and created variables using the GUI elements of the workbench, an easy way to preserve and recycle the effort is to export the experiment to CSV.   The first columns of the CSV file, preceding the data columns, will be the well name, sample name and the variables.   Thus they meet the requirements for being a sample name array.   Delete the data columns and dummy rows/columns.

To get started, you can download a template.

As an alternative to a sample sheet, a plate layout file may be more convenient.

See Also


Plate Layout

Help Front Page