The DUD-E paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3405771/ says that their decoys have similar physicochemical properties but different 2-D topologies in relation to the ligands.
¿What does it mean for a decoy to have a similar 2-D topology in relation to the ligand? ¿How a decoy differs structurally from a ligand?
I really appreciate all the help that will be provided!
hugs
Stellamaris
Hi Stellamaris,
Decoys are compiled as "inactive" molecules to a receptor. So these should have much higher binding free energies to a receptor, on the other hand, active molecules have lower ones. An ideal SBDD/LBDD program can tell active from inactive molecules exactly by some sort of metrics. But programs often give that decoys (inactive molecules) show better binding energies/scores to a receptor than those of active molecules, because decoys possess similar physicochemical properties to active molecules. So DUD-E dataset are used for benchmarking of SBDD/LBDD programs and/or their parameter settings. Try virtual screening with DUD-E.
Hi Stellamaris Ruth Sotomayor
Decoys: person or thing used to mislead or lure someone into a trap- means you want to check the performance of your model, like how good is your model to distinguish between the actives (experimentally proven) and decoys (misleading).
In drug design for the validation purpose, we use decoys to check the performance of the model like whether your model is able to distinguish between the actives and decoys (presumed inactives). When we select decoys, we select either on the basis of the molecular descriptor (number of donors, acceptors, number of rotatable bonds, etc), molecular weight, etc. Which means a decoy can be structurally different from the ligand.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology