Despite such benefits, gene expression data are frequently
peppered with missing values that may occur for a variety of
reasons. These include insufficient resolution, image corruption,
array fabrication error, and experimental error during the
laboratory processing [12]. Therefore, the treatment of missing
values is a critical step in preprocessing as data quality is a
major concern in consequent downstream analysis. This socalled
data cleaning [5] the need for repeating experiments,
which can be expensive and time consuming. Note also that the
repeat of experiments may not guarantee data completeness.