Consider a scenario where I have gone from a SMILES to a 3D structure (using openbabel). I then perform anyone of the openbabel finger print algorithms on that 3D structure before and after I run the structure through a classical force field energy minimisation. The finger prints are identical.
Are their fingerprint algorithms (and subsequent software), where a 3D conformation (defined by atom coordinates) make a difference to a drug's fingerprint?
Thanks
Anthony
Probably, it is because of the molecular fingerprints represent 2D structures in general (as example, Extended-connectivity fingerprints, a class of 2D fingerprint for molecular characterization).
Yes. The fingerprint is based entirely upon connectivity data - the 3D coordinates are not even considered.
Other types of descriptors often take 3D conformation into account, and in those cases, you'd expect a difference between the energy minimized and initial structures.
Yes, some encoding techniques exploit 3D information.
For example:
https://github.com/UnixJunkie/ACPC
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Nucleic Acids
- RNA