Download Knime extended by Omixon
Download 'KNIME installer with JRE6 and J3D', if there are no Java Runtime Environment 6 (JRE6) and Java 3D (J3D) on your computer.
Otherwise download only the missing software. (Note: The installation sequence is important: install firstly JRE6 , afterwards Java3D)
ChemInfoLogic-KNIME
ChemInfoLogic-KNIME is able to perform Tanimoto distance based similarity
calculations over molecule database.
There are three main calculation actions: Similarity search, Library to
library search and pharmacophore search,
all methods use binary fingerprint types to represent molecules.
The ChemInfoLogic-KNIME interface:
See the presentation videos:
Short description of the calculations:
Similarity search: Molecule or ligand fingerprinting
itself is a method that allows comparisons between ligands
to be made by representing chemical data of a molecule in numerical
form. Fingerprint generation algorithm examines
the molecule structure in a particular maximal bond length depth and
generates a fixed length binary string for
every molecule. The Tanimoto similarity score is a method of
calculating the similarity between two ligand
fingerprints. The result is a value between 0 and 1 with 0 indicating
no similarity and 1 indicating 100%
similarity. A high Tanimoto score of 0.7 or above, is representative of
two molecules having high structural
similarity which is a good indication of similar biological activity.
In this application the maximum size of
query database is 128 molecules. Program iterates over all query
molecules and calculates the Tanimoto distance
between the query molecule and one of the molecules of the reference
database, for every reference molecule.
Similarity search dialog box:
Similarity search output statistics:
Similarity search output histogram data:
Similarity search output histogram view:
Library to library: The comparison method
is same as in case of Similarity search action, the only difference is
that the query database is also a library, not only the reference
database.
Library to library search dialog box:
Library to library search 3D histogram view:
Pharmacophore search: Pharmacophore
fingerprints attempt to model binding related structural or chemical
properties of
chemical compounds with the use of simple statistics of chemical
features. This fingerprint contains prevalence information
about the ligands and atom groups which have pharmacological
significance. It has also fixed length. The similarity
measurement method is Tanimoto distance metric based as by Similarity
search action.
Pharmacophore search dialog box:
Pharmacophore search output:
Converting the output of Pharmacophore search in KNIME:
Pharmacophore search output after converting:
Beside the use of these functions, user needs to upload and modify
molecule databases and fingerprints belonging to them.
There are four main actions for this purpose:
User is able to upload SMILES and generate fingerprintsfrom them,
upload previously made fingerprints
(standard or also own types, if it fits the requirements of the given
calculation action),
update the previously uploaded libraries by a totally new fingerprint
file or by the enumeration of the
difference between the old and new library with the help of them.
Upload SMILES dialog box:
Upload fingerprints dialog box:
Update Library by New Version dialog box:
Update Library by Difference dialog box:
