ligand-based drug design

Ligand-based drug design is what you do when you cannot see the lock, so you study the keys that already work. Without a three-dimensional structure of the target, chemists infer what the binding site must look like by comparing many molecules that bind it, asking what features they share and what they lack.

From a set of known actives and inactives, you build a picture of the requirements for activity: which shapes, charges, and hydrogen-bonding groups matter, and where bulk is tolerated or forbidden. Tools include pharmacophore models that capture the essential interacting features, similarity comparisons, and quantitative structure–activity models that relate molecular descriptors to potency. The target's structure is invisible, but its 'negative space' is reconstructed from the ligands.

This approach is essential for targets that resist crystallography, such as many membrane proteins, and it can be very effective when good ligand data exist. Its weakness is that it cannot reveal anything the existing molecules never explored; it interpolates well within known chemistry but extrapolates poorly into genuinely new territory.