Solving structures

Why do scientists study proteins’ shapes?

To understand proteins fully we need to see what they look like, but even with a microscope they are too small to see. The key to ‘seeing’ the complex shapes of proteins was the discovery that pure proteins can be crystallised. This confirmed that they had a defined structure and meant that, in theory, their structures could be solved by X-ray crystallography. When you shine a beam of X-rays through a crystal, the regular array of protein molecules splits and diverts the beam in many directions, giving a pattern of spots that is recorded on a detector.

For simple crystals, such as diamond (which is not a protein), the maths needed to calculate the arrangement of atoms that makes a particular X-ray pattern is relatively straightforward. Proteins, with their thousands or even tens of thousands of atoms, presented a harder problem that took many years to solve. Today, big structures can often be solved using more intense X-ray sources and improved computational methods; however, not all proteins crystallise readily. Transmembrane proteins, in particular, are unstable when purified.

Protein structures are also probed with nuclear magnetic resonance, which is good for studying proteins in solution. Large protein complexes and ribosomes can be seen using electron microscopy.

One day, we may be able to predict how a protein folds from its amino acid sequence, but this is still a tough problem to solve because the number of possible shapes is astronomically large. Computer programmes have been developed to do some of the work, and scientists have tried to help them along by incorporating them into a computer game that anyone can play – try it!

About this resource

This resource was first published in ‘Proteins’ in January 2014.

Cell biology, Biotechnology and engineering
Education levels:
16–19, Continuing professional development