Recognizing Pitfalls in Virtual Screening: A Critical Review

By T. Scior, Andreas Bender, G. Tresadern, José L. Medina-Franco, Karina Martínez-Mayorga, T. Langer, Karina Cuanalo-Contreras, D. Agrafiotis. 2012

394 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Molecular Biology, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Molecular Biology

In Computational Chemistry & Drug Discovery

Citations in the corpus

Molecular Docking and Structure-Based Drug Design Strategies (2015, background): “Rigorous review of the available information regarding the target and known ligands, as well as a careful analysis of the advantages and pitfalls of the selected docking algorithms, are required in delineating the most appropriate strategies [169].”
Effects of multiple conformers per compound upon 3-D similarity search and bioassay data analysis (2012, background): “[12], to increase the reliability of the 3-D similarity comparison one must consider as many conformations per compound as possible to ensure adequate conformational space coverage.”
PubChem3D: conformer ensemble accuracy (2013, background): “[53], because adequate conformational space coverage is an important requirement for reliable 3-D similarity computations, it would be desirable to consider as many conformations per molecule as possible.”
Benchmarking platform for ligand-based virtual screening (2016, background): “The top ranked molecules can then be passed to biochemical screening [7].”
Open-source platform to benchmark fingerprints for ligand-based virtual screening (2013, background): “The performance of a fingerprint in these studies depends not only on its ability to describe the molecular features responsible for activity against a specific target, but also on the composition of the data set, the statistical robustness of the study, and the evaluation method(s) [9-14].”
Performance of machine-learning scoring functions in structure-based virtual screening (2017, background): “1) (1) per-target – in this approach, we generate 102 unique machine-learning SFs, each created independently for a single protein target (trained only on its active and decoys ligands); (2) horizontal split – both training and test sets contain data from all targets, i.”
Molecular fingerprint similarity search in virtual screening. (2015, methodology): “• Conformation coverage: When using 3D fingerprints, the conformations for each molecule should adequately cover its conformational space, which requires the testing and optimization of several parameters [70].”