The SSMD (strictly standardized mean difference) is a statistical parameter that describes the quality of an assay or a screening plate. It takes into account the separation and the variabilities of two controls (e.g. positive and negative control wells) and ranges from plus to minus infinity. The SSMD as an alternative to the Z-Prime factor is a bit easier to interpret because it behaves similar as the Z score.

For groups of data (e.g. wells of plates) SSMDs are calculated for two selected subsets (e.g. positive and negative control wells of a plate). Multiple positive and negative controls can be selected as subsets and the SSMDs will be calculated for all combinations of these. For each group mean and standard deviation (or median and mad) are calculated for the selected subsets.

SSMD = (mean(x[subset{positive}]) - mean(x[subset{negative}])) / sqrt(sd(x[subset{positive}])² + sd(x[subset{negative}])²)
SSMD = (median(x[subset{positive}]) - median(x[subset{negative}])) / sqrt(mad(x[subset{positive}])² + mad(x[subset{negative}])²)

Literature: Zhang, X., Genomics 89 (2007) 552–561
Wikipedia: SSMD


Group wells by
Select the column to define the groups (e.g. "barcode" for plate-wise SSMDs).
Subset column / well annotation column
Column with well annotations (e.g. treatment)
Subset / Positive Control
Select subsets as positive controls for which means and sd (medians and mads) are calculated for each group.
Subset / Negative Control
Select subsets as negative controls for which means and sd (medians and mads) are calculated for each group.
Use robust statistics
Median and mad will be calculated instead of mean and sd.
Select the columns for which SSDMs will be calculated.

Input Ports

Input table

Output Ports

Table with SSMDs


This node has no views




You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.