Normalization workflow processes raw plate data in a matrix format and performs four commonly used HTS data normalization techniques:

**Z-score, Robust Z-score, Median polish and B-score**. Results are written in both simple text (*csv*) and excel (*xls*) file formats. The normalizations are implemented using the R programming environment with the R snippet Knime nodes.
Hit identification workflow processes raw plate data in a matrix format and calculates intraplate

**standard deviations**(deviation from plate mean) and**median absolute deviations**(deviation from plate median). These values then serve for primary HTS hit identification. The hits are selected on the basis of user defined cutoff values (parameters*k*and*l*). Results are written in both simple text (*csv*) and excel (*xls*) file formats. For both formats there are two files: one with binary matrix representing original plate, where 1 stands for hit and 0 for inactive well; second with coordinate of wells with hits. Also a scatter plot image (png) with highlighted hits is exported together with the text exports.
Chemical space workflow visualizes molecules distance/similarity in a 2D space using 2 most common dimensionality reduction techniques:

**Principal Component Analysis (PCA)**and**Multidimensional Scaling (MDS)**.
Molecules are read from a

*SDF*file and a set of chosen descriptors*(e.g., molecular weight, LogP, number of rotatable bonds etc.)*is calculated using*RDKit Descriptor Calculation node*. Then in one branch of the workflow, these descriptors are used to calculate the principal components of the set. In the second branch,*Morgan fingerprints with radius 3 and 1024 bits*(hashed binary fingerprints) are also calculated for the set and projected in 2D space by MDS.
In both cases the result of projection is visualized using a scatterplot connected with

*a HiLite Row Splitter node*. This node is able to interactively show selected points (molecules) via a 2D/3D Scatterplot node. In the Scatterplot node, the visualized points can be colored and sized by one of the descriptors (dimensions).