Does a good docking score guarantee a strong binder?

No; scoring functions are approximate and prone to false positives, so docking is best used to enrich and prioritize candidates rather than to predict affinities precisely without experimental confirmation.

What is the difference between docking and virtual screening?

Docking predicts how one ligand binds a target, while virtual screening applies docking or other models across a large library to select which compounds to test.

Molecular Docking and Virtual Screening

Molecular docking predicts how a small molecule binds to a target, and virtual screening applies this and related methods to rank large libraries of candidate compounds.

Definition

Computational methods that predict the bound conformation and relative affinity of ligands to a macromolecular target and use these predictions to prioritize candidate molecules.

Scope

Covers the docking problem of predicting binding pose and affinity, conformational search algorithms, scoring functions and their limitations, and structure-based and ligand-based virtual screening of compound libraries. Centers on computer-aided drug design applications.

Core questions

How is a ligand's binding pose searched and predicted within a target site?
How do scoring functions estimate binding affinity, and why are they imperfect?
How does virtual screening triage large libraries efficiently?
When are structure-based versus ligand-based approaches appropriate?

Key theories

Pose generation and scoring: Docking decomposes binding prediction into searching plausible ligand poses and scoring them with an approximate function balancing accuracy against the speed needed to screen many molecules.
Virtual screening triage: Computational ranking filters large libraries to a tractable set of promising candidates for experimental testing, using structure-based docking or ligand-based similarity.

Clinical relevance

Docking and virtual screening are core tools of structure-based drug discovery, helping to identify and prioritize hit and lead compounds and to rationalize binding, thereby focusing costly experimental campaigns.

History

Beginning with Kuntz's DOCK program in the early 1980s, docking matured alongside growing structural databases and computing power; scoring functions and virtual screening protocols became central to pharmaceutical discovery from the 1990s onward.

Debates

Reliability of scoring functions: Scoring functions trade physical rigor for speed and often rank actives only modestly better than chance for affinity, so their predictive reliability and best validation practices remain debated.

Key figures

Irwin Kuntz
Jürgen Bajorath
Andrew Leach
Brian Shoichet

Seminal works

kitchen2004

Frequently asked questions

Does a good docking score guarantee a strong binder?: No; scoring functions are approximate and prone to false positives, so docking is best used to enrich and prioritize candidates rather than to predict affinities precisely without experimental confirmation.
What is the difference between docking and virtual screening?: Docking predicts how one ligand binds a target, while virtual screening applies docking or other models across a large library to select which compounds to test.