The features represent subgraphs of the input molecule. For a binary fingerprint they are hashed (ireversally) and there is a many-to-one mapping between the subgraph and the hash (i.e. the value 14 you see). We do not currently provide a general way to see what features hash to which values but some fingerprints have an option to generate the features "unfolded". We should add this option in more places since it can be useful.
Best Wishes, John On Mon, 28 Aug 2023 at 19:06, Chong Kim San Allen via Cdk-user < cdk-user@lists.sourceforge.net> wrote: > Dear Helpdesk, > > > > I have used CDK to generate the Extended Fingerprints for a couple of > compounds and I found that certain features are common among my compounds. > For example, “14” keeps showing up. I would like to know what is “14”? I > know that the default path length is 7 so I was wondering if the feature is > a chemical substructure? The default size for Extended Fingerprint is 1024 > so I was wondering if there is a way to figure out what each of the 1024 > features represents. > > > > Similarly, if I generated ECFP6 which has 2^32 features (count version), > is there a way for me to figure out what each of those features are? If a > feature appears to have a high count and I wanted to figure out what this > feature was, is there a command I can use to find out what that feature > represents? > > > > Thanks in advance for your help. > > > > Best, > Allen > > > ------------------------------ > > CONFIDENTIALITY: This email is intended solely for the person(s) named and > may be confidential and/or privileged. If you are not the intended > recipient, please delete it, notify us and do not copy, use, or disclose > its contents. > Towards a sustainable earth: Print only when necessary. Thank you. > _______________________________________________ > Cdk-user mailing list > Cdk-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/cdk-user >
_______________________________________________ Cdk-user mailing list Cdk-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/cdk-user
