Preprocessing / Impute missing values
Description
Fills in the missing values in the data by estimated values.
Parameters
- Imputation method (mean, median, knn) [knn]
- Row wise or column wise (column, row) [column]
- Number of neighbor (1....10000) [5]
- Maximum number of missing values on a row (0...100) [20]
- Maximum number of missing values on a column (0...100) [20]
Details
The missing values can be replaced with the mean or median of the array/sample (column) or genes (rows), or they can be
estimated using the specified number of closest neighbors (knn). If the maximum number
of genes missing either on a row or on a column is larger than the selected number, the missing
values will not be estimated. This is implemented in order to protect users from imputing too
large a fraction of their dataset, since this will probably affect the result tragically.
Output
A tabular text file with intensity values for the genes.