How can we find disimilar compounds in the database

Sir,

We have a library of 50,000 compounds. Now we want to find disimilar compounds from the library of 50,000. How can we find dissimilar compounds..Please reply its urgent

Seerat

 

Seerat,

You may run overlap analysis on the table and do the similarity search. Target and query table will be the same in this case. Please read the documentation at https://www.chemaxon.com/instantjchem/ijc_latest/docs/user/help/htmlfiles/chemistry_functions/performing_overlap_analysis.html.

Hope it helps,

Filip 

Hello Sir

Greetings of the day!

Thanx for the reply.

But my query is that i want to find the disimilar compounds from the list of 50,000 ...means instant jchem itself  extract the disimilar compounds from the whole list

Seerat

Good morning, Sir.

After performing overlap analysis there is a new column called Overlap count. Basically the compound which have 0 in this field are unique for the defined similarity treshold. These are the most dissimilar coumpounds in the set.  

The approach would be to search for the compounds which have 0 overlap count. The results can be then exported to a file or you can make a list from it.

Regards,

Filip

Would there be a way to select one representative per cluster with this method ?

The only way I know of doing such sofar is by using knime.

Ellert,

there is currently no way how to select only one most distinct compound in a cluster with this method.

Filip