ADME and drug-likeness: Lipinski's rule of five explained

A plain-language guide to ADME, the drug-likeness rules used in computational screening and their limits.

What ADME stands for

ADME stands for absorption, distribution, metabolism and excretion. It describes what the body does to a compound. Adding toxicity gives ADMET. Poor ADME is a major reason that otherwise potent compounds fail.

Common drug-likeness rules

Item	What it is	How to read it
Lipinski (rule of five)	Molecular weight ≤ 500, logP ≤ 5, H-bond donors ≤ 5, acceptors ≤ 10.	Oral absorption likelihood. One violation is often tolerated.
Veber	Rotatable bonds ≤ 10 and polar surface area ≤ 140 Å².	Oral bioavailability in rats, used as a flexibility and polarity check.
Ghose	Molecular weight 160 to 480, logP −0.4 to 5.6, atoms 20 to 70.	A broader range based on known drugs.
Egan and Muegge	Polar surface area and logP limits, plus other property bounds.	Alternative filters for absorption.

How to use ADME predictions

Standardise and neutralise your structures first.
Calculate descriptors and apply the rule sets.
Predict solubility, permeability and CYP interaction likelihood.
Flag structural alerts that suggest reactivity or toxicity.
Combine with docking so top-scoring compounds also look drug-like.

Limits to keep in mind

Rules are guidelines. Many successful drugs, especially natural products, break them.
Predictions have errors and are weakest outside the chemical space of their training data.
In silico ADME never replaces experimental testing.

Frequently asked questions

What is Lipinski's rule of five?

A guideline that oral drugs tend to have molecular weight under 500, logP under 5, no more than 5 hydrogen-bond donors and no more than 10 acceptors.

Is it a problem if a compound breaks one rule?

Not necessarily. Lipinski's guideline is usually applied with tolerance for one violation, and some drug classes lie outside it.

What is the difference between ADME and ADMET?

ADMET adds toxicity to absorption, distribution, metabolism and excretion.

Can ADME be predicted before synthesis?

Yes, as estimates from structure. They help prioritise compounds but do not replace measurements.

Prepare the protein and ligands, define the binding site, validate by re-docking, dock your compounds, then inspect interactions instead of trusting scores alone.

Virtual screening guideVirtual Screening Workflow: From Library to Hit List

Prepare the target and library, filter by properties, validate the docking setup, dock, then rank, cluster and inspect before choosing compounds to test.

Related service

ServiceADME prediction

In silico ADME and drug-likeness prediction: Lipinski and Veber rules, physicochemical descriptors, solubility, permeability and toxicity alerts for compound prioritisation.

Written by Sohel Akhtar Ansari, ReactorMD founder (ORCID). Computational predictions are not experimental confirmation.