Differences

This shows you the differences between two versions of the page.

--- diversity_selection [2012/07/02 20:32] – [Basic options] rkiss
+++ diversity_selection [2012/07/03 17:42] – sanmark
@@ Line 16: / Line 16: @@
   * **Similarity threshold**: the maximum S similarity allowed in the diverse set. It is guaranteed that none of the resulting molecules are more similar than S.
-  * **Number of most diverse molecules**: the maximum N number of diverse molecules to be selected. The diversity selection algorithm will select the most diverse N molecules, unless the maximum allowed similarity is reached first.
+  * **Max number of most diverse molecules**: the maximum N number of diverse molecules to be selected. The diversity selection algorithm will select the most diverse N molecules, unless the maximum allowed similarity is reached first.
-If you do not limit the selection, the full collection will be returned ordered by diversity. This means that the top N molecules in the resulting collection will be the most diverse N molecules. The //maximum similarity// found at the Nth molecule in the results refer to the minimum diversity of the first N molecules (none of them are more similar than that value).
+If you do not limit the selection, the full collection will be returned ordered by diversity. This means that the top N molecules in the resulting collection will be the most diverse N molecules. The //maximum similarity// found at the Nth molecule in the results will refer to the minimum diversity of the first N molecules (none of them are more similar than that value).
 ==== Advanced options ====
-You can adjust the meaning of similarity and dissimilarity of molecules here, selecting the descriptor on which similarity scores are calculated. We use the Tanimoto coefficient (Jackard index)((http://en.wikipedia.org/wiki/Jaccard_index)) as the measure of similarity now, but you can chose between different chemical fingerprints as descriptors. We plan to introduce more descriptor types and more similarity measure types in the future.
+Under Advanced options, you can adjust the definition of similarity/dissimilarity of molecules. You can select the descriptor used for calculating the similarity scores. Currently two fingerprints (OpenBabel Linear Fingerprint and Indigo Similarity Fingerprint) are available.
-  * **Molecular descriptor**: the molecular descriptor used to represent chemical structures during the calculation
+You will be able to set different similarity metrics as the measure of similarity. Currently, only the Tanimoto coefficient (Jaccard index)((http://en.wikipedia.org/wiki/Jaccard_index)) is implemented as the measure of similarity.
-==== Default options ====
+We plan to introduce more descriptors and more similarity measure types in the future.
-The default descriptor used is the linear fingerprint implemented in Open Babel ((Open Babel v2.3.90 http://openbabel.sourceforge.net/)), which is similar to Daylight’s fingerprint and Chemaxon’s linear fingerprint, and the Tanimoto coefficient is calculated as the similarity of fingerprints.
+  * **Molecular descriptor**: the molecular descriptor applied for representing chemical structures during the calculation
-If you have no suggestions to use another setup, you can rely on our choices. After implementation and evaluation of new fingerprints and metrics, the default setup can be changed. This can be tracked at the end of this document, in the Changelog section.
+==== Default options ====
+The default descriptor used is the linear fingerprint implemented in OpenBabel ((Open Babel v2.3.90 http://openbabel.sourceforge.net/)), which is similar to Daylight’s fingerprint and ChemAxon’s ((http://www.chemaxon.com/jchem/doc/user/fingerprint.html)) linear fingerprint, and the Tanimoto coefficient is calculated as the similarity of fingerprints.
 ==== Algorithm ====
-We use an optimized implementation of the stepwise elimination algorithm((R. J. Taylor, J. Chem. Inf. Comput. Sci., 1995, 35, 59 67.)), which can be described as follows:
+We use an optimized implementation of the stepwise elimination algorithm((R. J. Taylor, J. Chem. Inf. Comput. Sci., 1995, 35, 59-67.)), which can be described as follows:
-  - calculate the similarity matrix of the molecules in the input collection
+  - Calculate the similarity matrix of the molecules in the input collection
-  - process the matrix elements as follows:
+  - Process the matrix elements as follows:
-    - select the largest off-diagonal element in the similarity matrix
+    - Select the largest off-diagonal element in the similarity matrix
-    - eliminate one molecule of the most similar molecule pair randomly
+    - Eliminate one molecule of the most similar molecule pair randomly
-    - go to step I. if off-diagonal elements remained
+    - Go to step I. if off-diagonal elements remained
-  - sort the list of eliminated molecules by similarity values associated to the elimination steps in increasing order
+  - Sort the list of eliminated molecules by similarity values associated to the elimination steps in increasing order
-During this process, the size of the collection is reduced and diversity increases. Each elimination step throws out a compound that has close analogues in the remaining set. In result, we get a single compound, and a list of compounds with decreasing similarity values, which can be interpreted as the increasing diversity of the remaining set.
+During this process, the size of the collection is reduced while the diversity of the collection is increased. Each elimination step filters out one molecule that has close analogues in the remaining set. As a result, the remaining molecules will have a decreased similarity (increased diversity).
-After the algorithm finishes, structures are sorted by similarity values and are placed in the result collection. The first molecules in the resulted collection are the most dissimilar (most diverse) ones, and the length of the result list is determined using the diversity options: the maximum number of compounds to be selected and the maximum similarity values allowed between diverse compounds.
+After the algorithm finishes, structures are sorted by similarity values and are placed in the result collection. The first molecules in the resulted collection are the most dissimilar (most diverse) ones. The length of the result list is determined by input parameters: maximum number of compounds and similarity threshold.
 ===== Limitations =====
-The diversity selection is freely accessible for every mcule user with a monthly limit of 10000 input compounds. The average run time for 10000 compouds is about 5 minutes. The usage of your diversity filter can be tracked on the user profile / limits. Our technologies allow effective processing of very large collections (~10M). If you want to exceed your limits, please contact us.
+Diversity selection filter is freely accessible for registered mcule users. Monthly limit is set to 10,000 input molecules. Your usage limits including the number of remaining molecules can be tracked under user profile / limits. To check your user profile click on your user name in the upper right corner on the mcule.com website. If you would like to run the Diversity selection filter for larger collections, please [[support@mcule.com|contact us]]. Our technology allows effective processing of very large collections (~10M).
+The average run time for 10,000 input molecules about a minute.
 ===== Changelog =====