SAR by synthesis
There are two ways to figure out how something works: stare at it, or take it apart and rebuild it piece by piece. SAR by synthesis is the hands-on approach to understanding a molecule: you actually make analog after analog, test each one, and let the pattern of results tell you which structural features matter.
In practice, SAR by synthesis is an iterative loop. A hypothesis suggests which positions to vary; chemists synthesize a focused set of analogs probing those positions; biological assays measure each compound; the results refine the hypothesis; and the cycle repeats. Over many rounds this maps out which features are essential, which are tunable, and which are dead weight, gradually defining the molecule's binding requirements through real, tested compounds rather than predictions.
This empirical, build-and-measure strategy is the historical backbone of medicinal chemistry and remains indispensable, especially when no protein structure is available to guide design. Its cost is exactly its strength: every data point requires real synthesis and assay time, so SAR by synthesis is powerful but slow and resource-hungry, which is why computational and structure-based methods are used alongside it to choose which analogs are worth making.
Lacking a crystal structure, a team makes forty analogs varying three regions of a hit; the survivors that retain potency reveal which region tolerates change and which must stay fixed.
Knowledge is bought one synthesized, tested compound at a time.
SAR by synthesis is the experimental engine inside the broader design–make–test cycle; computational design and structure-based methods help decide which analogs to feed into it.