Hidden bias in the DUD-E dataset leads to misleading performance of deep learning in structure-based virtual screening.
Recently much effort has been invested in using convolutional neural network (CNN) models trained on 3D structural images of protein-ligand complexes to distinguish binding from non-binding ligands for virtual screening. However, the dearth of reliable protein-ligand x-ray structures and binding aff...
Main Authors: | Lieyang Chen, Anthony Cruz, Steven Ramsey, Callum J Dickson, Jose S Duca, Viktor Hornak, David R Koes, Tom Kurtzman |
---|---|
Format: | Article |
Language: | English |
Published: |
Public Library of Science (PLoS)
2019-01-01
|
Series: | PLoS ONE |
Online Access: | https://doi.org/10.1371/journal.pone.0220113 |
Similar Items
-
Pneumonia: Bugs, Drugs, and Laboratory Duds
by: Accardi, Andrew MD
Published: (2003-04-01) -
CALCULATION OF PER PARCEL PROBABILITY FOR DUD BOMBS IN GERMANY
by: S. M. Tavakkoli Sabour, et al.
Published: (2014-10-01) -
AL-'ADAD WA AL-MA'DUD DALAM ALQURAN
by: M. Jabir
Published: (2006-06-01) -
Stud or Dud? Representations of masculinity in the popular culture of 1950s America
by: Ruckley, Emma
Published: (2008) -
Article 7: Why the 'Nuclear Option' turned out to be a dud
by: Dethlefsen, Mads Schou
Published: (2019)