meaning of distance matrix output

KNIME Analytics Platform
1

hello!

I´m using the distance matrix node in my workflow. I calculate fingerprits and then i use the tanimoto function of the distance matrix node calculating a distance matrix of the fingerprints.

now in the output column there are the distances listed but i don´t understand between which row values the distance is meaned?! - ok, in the first row there is no distance because it compares the first entry with itself. in the second row there is the distance between the first and the second entry.. ok, but what is the first entry of the third row? is it the distance between molecule 3 and 1 or between 3 and 2? and waht is then the second entry of the third row (distance between which  row entries)? and so one...

thanks for any reply!

2

it is a lower triangular distance matrix.

 

e.g. in DataRow i at position j you find the distance between the molecule in Row i and the molecule in Row j (only for i<j)

3

Hi

If you dont want to a matrix of all the molecules being compared with one another and simply want to get a tanimato similarity between one molecule (or a set of molecules) and a list of molecules then you can use either the Indigo Fingerprint Similarity or the Erlwood Fingerprint Similarity.

If you are using more than one molecule to compare against, and want an overall average similarity then simply tick Multifusion query in ther Erlwood node, or average aggregation type in the Indigo node.

Simon.