Lipinski's Rule of Five: Screening Drug Candidates by Molecular Properties

The Oral Bioavailability Bottleneck

By the 1990s, pharmaceutical companies had automated the process of building new candidate molecules through combinatorial chemistry and were synthesizing tens of thousands of compounds a year, far more than the small number that could ever be pushed through animal studies and clinical trials. The bottleneck was not finding molecules that could bind tightly to a biological target in a test tube; that part was getting easier. The bottleneck was oral bioavailability: the fraction of an orally swallowed dose that survives digestion, crosses the intestinal wall, and reaches the bloodstream intact. A molecule could be a spectacular binder in a petri dish and still fail completely as a drug if it was too large, too greasy, or too studded with polar groups to pass through the lipid membranes lining the gut. Lipinski, working at Pfizer, analyzed the physical properties of roughly 2,300 drugs that had reached at least Phase II clinical trials and looked for the numeric patterns that separated compounds with decent oral absorption from compounds that never made it past preclinical testing. The result was not a rigorous physical law but a simple, memorable filter that could be calculated instantly from a molecule's structure, letting chemists throw out obviously doomed candidates before spending months and real money on wet-lab testing.

The Four Thresholds and What They Mean Physically

Lipinski's Rule of Five sets four thresholds, each of which flags a molecule as a poor oral drug candidate if it is violated. First, molecular weight (MW) is less than or equal to 500 Daltons; heavier molecules are generally too bulky to diffuse efficiently across the tightly packed lipid bilayer of intestinal epithelial cells. Second, the calculated octanol-water partition coefficient, LogP, is less than or equal to 5; LogP measures how a molecule distributes between a fatty (octanol) phase and a watery phase, and it needs to be greasy enough to slip through a lipid membrane but not so greasy that it never dissolves in the watery fluids of the gut and blood in the first place. Third, the number of hydrogen bond donors is less than or equal to 5, counted as the sum of all OH and NH groups on the molecule; each donor group forms strong hydrogen bonds with water that must be stripped away before the molecule can pass into the oily membrane interior, so too many donors act like anchors holding the molecule in the watery phase. Fourth, the number of hydrogen bond acceptors is less than or equal to 10, counted as the sum of all nitrogen and oxygen atoms in the molecule, since these atoms also pull in water molecules through hydrogen bonding and add to that same membrane-crossing penalty. Every one of these four numbers, 500, 5, 5, and 10, is a multiple of 5, which is exactly why chemists nicknamed it the Rule of Five: an easy mnemonic for an easy calculation.

Why More Than One Violation Is a Red Flag

Lipinski's original 1997 analysis found something important: a single violation of any one threshold did not, by itself, doom a compound's oral absorption. Plenty of successful oral drugs sit just over the LogP limit of 5, or carry a molecular weight of 520 or 540 Daltons, and still absorb reasonably well because the other three properties compensate. But when a molecule violated two or more of the four rules simultaneously, the probability of poor absorption or poor permeability rose sharply. The logic is additive: each violated threshold represents another physical obstacle, whether it is excess bulk, excess polarity, or too many water-hungry hydrogen-bonding groups, and intestinal membranes are unforgiving to molecules stacking up multiple obstacles at once. This is why the rule is almost always applied as a counting exercise rather than four independent pass/fail gates: a compound with MW of 480, LogP of 4.2, 3 hydrogen bond donors, and 7 hydrogen bond acceptors passes cleanly on all four counts, while a compound with MW of 560 and LogP of 5.8 (two violations) is flagged as a high-risk candidate for absorption problems, even though each individual number is only modestly over its threshold.

A Fast First-Pass Filter in Modern Drug Discovery

Today, Lipinski's Rule of Five is baked into the software used at essentially every pharmaceutical and biotech company as a near-instant computational filter. Because MW, LogP, hydrogen bond donor count, and hydrogen bond acceptor count can all be calculated directly from a molecule's two-dimensional structure in milliseconds, cheminformatics pipelines routinely screen virtual libraries of millions of candidate structures and discard the ones with two or more rule violations before a single physical sample is made. This matters enormously for cost and speed: synthesizing a compound and running it through absorption assays can take days to weeks and real laboratory budget, while a Rule of Five check takes essentially no time at all. The rule is typically used alongside other computed properties, such as topological polar surface area and rotatable bond count, as part of a broader drug-likeness or ADMET (absorption, distribution, metabolism, excretion, toxicity) screening funnel, where cheap computational filters progressively narrow enormous candidate pools down to a manageable shortlist worth the expense of actual synthesis and testing.

Limitations: A Guideline, Not an Absolute Law

Lipinski himself was careful to describe the Rule of Five as a guideline drawn from a specific dataset, not a hard physical law, and its exceptions are numerous and well known. Natural product-derived drugs, antibiotics, and many antifungal agents routinely violate multiple thresholds; for example, the antibiotic vancomycin has a molecular weight well over 1,400 Daltons and dozens of hydrogen bond donors and acceptors, yet remains clinically useful, in part because such molecules absorb through mechanisms other than simple passive diffusion, including active transport by membrane proteins. The rule was also explicitly designed only for small, orally dosed, passively absorbed molecules, so it does not apply to biologics such as antibodies, peptides, and vaccines, which are administered by injection and bypass the intestinal barrier entirely, or to drugs that rely on active transporters to cross membranes. Because of these blind spots, modern drug discovery treats a Rule of Five violation as a prompt for closer scrutiny rather than an automatic rejection, and later refinements such as the Rule of Three for fragment-based drug design, or Veber's rules incorporating rotatable bonds and polar surface area, have extended and modified Lipinski's original four thresholds for different classes of chemistry.

Frequently asked questions

Try it live

Everything above runs in your browser — open Lipinski's Rule of Five: Screening Drug Candidates by Molecular Properties and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.