For every term in the categorical columns it is tested whether the corresponding expression values have a distribution in two-dimensional planes of expression values that is deviating from the global distribution. For details see Cox and Mann (2012) BMC Bioinformatics 13 Suppl 16:S12.
Output: The output matrix contains a list of terms from all categorical columns that are significantly biased compared to the global distribution.
Selected numerical/expression columns that are used as x-axis for the testing of the 2D distributions (default: no expression/numerical columns are selected).
Hint: The selected number of columns in “Columns1” have to be equal to the selected ones in “Columns2”.
Selected numerical/expression columns that are used as y-axis for the testing of the 2D distributions (default: no expression/numerical columns are selected).
Hint: The selected number of columns in “Columns2” have to be equal to the selected ones in “Columns1”.
The truncation can be based on p-values or the Benjamini-Hochberg correction for multiple hypothesis testing (default: Benjamini-Hochberg FDRFalse Discovery Rate). Rows with a test result below a specified value (“Threshold value”) are reported as significant.
Based on a specified threshold (default: 0.02) a row is reported as significant, if its test result is below the defined value. Depending on the chosen truncation score this threshold value is applied to the p-value or to the Benjamini-Hochberg FDR.
Selected text column, where all rows having the same identifier will be counted as one entity in the 2D annotation enrichment analysis (default: <None>). The main application is for posttranslational modification sites. Then one should select here protein or gene identifiers. This will make sure that multiple sites from the same protein (or gene) are counted only once for the enrichment analysis.