What is a drug target, really?

A medicine is not magic and it is not general. Almost every drug you have ever taken works by physically sticking to one specific molecule inside your body and changing what it does.

That molecule is called the target.

The lock and the key

The usual comparison is a lock and a key, and it is a decent one as long as you remember which is which. The target is the lock. The drug is the key.

Most targets are proteins — molecules your cells build, folded into specific three-dimensional shapes. The shape is the whole point. A protein has grooves and pockets on its surface, and a drug works by being exactly the right shape to settle into one of them.

When a drug is sitting in that pocket, the protein cannot do its normal job. Usually that is the goal: the protein was doing something you want stopped.

Ibuprofen works this way. It blocks an enzyme called COX, which your body uses to make the molecules that produce pain and swelling. Block the enzyme, make fewer of those molecules, feel less pain. It is not treating the injury. It is interrupting one specific step.

Why choosing the target is the hard part

Here is the thing people outside the field tend to miss. Making the key is difficult, but it is a problem chemists are very good at. Choosing which lock to pick is the decision that actually determines whether the medicine works.

Get it wrong and everything downstream is wasted — years of chemistry, years of testing, hundreds of millions of dollars, all aimed at a protein that turned out not to matter much for the disease.

And you often cannot tell you were wrong until very late. A drug can work beautifully in a dish, work in mice, and then fail in people, because the target was never really driving the human disease in the first place.

This is why evidence about targets is worth so much. Before anyone makes a single molecule, you want the strongest possible case that this protein, in actual humans, actually matters.

Not every target can be drugged

One more complication. Some proteins have nice deep pockets that a small molecule can nestle into. Those are called druggable.

Others are smooth. Nothing to grip. Or they are structural — bricks in a wall rather than switches on a machine — and there is no switch to flip. Scientists call these targets undruggable, though that word is doing a lot of work. It usually means "nobody has managed it yet," and the list gets shorter every decade.

So a target needs two separate things to be useful: strong evidence that it matters in the disease, and some realistic way to reach it. Plenty of proteins have one without the other.

Try this

1. Go to ChEMBL, a free database of drugs and their targets: https://www.ebi.ac.uk/chembl/
2. Search for a medicine you have actually heard of. Try ibuprofen, or loratadine if you take anything for allergies.
3. Find the Mechanism of Action section. It names the exact protein the drug sticks to.
4. Now search that protein and see how many other drugs target the same thing.

That last number is often surprising. Once somebody proves a target works, everyone builds a key for the same lock.

One question to take away

If a protein has excellent evidence behind it but no pocket to grip, should scientists keep working on it — or move to an easier target with weaker evidence?

Real research teams argue about this constantly, and there is no settled answer.